Origin of The Species and Genus Concepts

Origin of the Species and Genus Concepts:
An Anthropological Perspective
SCOTT ATRAN
C.N.R.S. UA 882
Laboratoire d'Ethnobiologie-Biogdographie,
Museum National d'Histoire Naturelle
Paris, France
In his dedicatory epistle to the founding work of systematic

botany, A n d r e a Cesalpino disclaims the practices of earlier natural
historians who, like Dioscorides,
assembled [plants] by medicinal properties . . . . Others have
classified alphabetically, so that one can more easily store
content in memory . . . . But the classification that unites plants
by their natural affinities may be considered the easiest, the
surest and the most efficacious, be it for the memory or for the
study of [a plant's secondary] properties.
In a natural system, then,
genera and species are constituted neither from medicinal
properties, nor by reason of some use, nor from the place in
which they occur. These are all accidents. (Cesalpino 1583:26)
By attempting to proceed straightaway to an account of the
various virtues of natural kinds, without first carefully describing
those kinds in regard to morphological similarities and differences,
ancient and medieval herbalists succeeded only in confounding
our spontaneous appreciation of natural relationships; "It seems,"
notes Tournefort, "that the more they enriched medicine, the more
they threw botany into confusion" (16 94:7 ).
What is curious here is the suggestion that ancient natural
history angled away from intuition. It is odd because Cesalpino
and Tournefort did not presume that men and women before them
had successfully classified plants or animals in accordance with
such supposedly apparent intuitions -- what Linnaeus (1715, sec.
168) would intend by p r i m o intuitu e x f a c i e externa. If the learned
scholars of Greece, Rome, and the Renaissance ultimately failed
on this score, then surely it would be folly to suppose that more
primitive ancestors or contemporary preliterate peoples could
Journal of the History of Biology, Vol. 20, No. 2 (Summer 1987), pp. 195--279.
1987 by D. Reidel Publishing Company.
196
SCOTT ATRAN
do better - - the unspoiled savage of Montaigne notwithstanding.

T h e r e are, it is true, significant resemblances to be f o u n d
between the ancient herbals of G r e e c e and R o m e , their medieval
and Renaissance derivatives, and those of Mesoamerica, the
Middle East, China, and India. Frequently there a p p e a r to be
"unnatural" groupings of plants - - that is, groupings based not on
any readily perceived morphological similarity, but on some
a c k n o w l e d g e d virtue. In E u r o p e a n herbals since Pliny, for instance, various species of E u p h o r b i a c e a e are often g r o u p e d with
plants of other families because of a similarity in the texture and
color of the sap, which was valued as a medicinal purgative.
Similarly, in the sixteenth-century A z t e c herbal k n o w n as the
Badianus Manuscript ( E m m a r t 1940), two illustrated plants appear: the tohmioxihuitl ("hairy plant"), belonging to the tribe
Chichorieae of the Compositae, and memeyaxiuhtontli ("little milk
plant"), a species of Euphorbia. T h e s e are also linked together on
the basis of their milky juice - - a juice that allegedly increases
lactation in women.
It would be a mistake, however, to conclude, as so m a n y have
d o n e and continue to do f r o m observations such as these (cf. Li
1974; M o r t o n 1981:1; M a y r 1982:134), that knowledge of living
kinds was exclusively, or even primarily, a practical knowledge
d e p e n d e n t on cultural virtues. ~ Admittedly, the initial task of
ancient natural historians, or rhizotomists, was to provide a written
record of those features of plants (and animals) that might be
d e e m e d beneficial or harmful to man's well-being, especially the
"hidden" aspects of roots, sap, and entrails. But such functional
knowledge was itself d e p e n d e n t on a rich and prior c o m p e t e n c e
with respect to the visible patterns of the living world. It does indeed
1. The cultural factors allegedly responsible for the original forms of folkbiological classification are thought to run the gamut from the strictly economic
to the purely symbolic (cf. Croizat 1945; Gilmour and Waiters 1964). According
to Li, (1974) the beginnings of classification were inextricably bound to the
practice of cultivation. Studies of contemporary Neolithic societies, though,
indicate this is not so. Morton claims that Theophrastus's fourfold division of the
plant kingdom into tree, shrub, undershrub, and herb "was essentially a tribal
classification, reflecting the most primitive, originally functional, division of the
tribe into two, later four, and then more numerous moieties" (Morton 1981:1).
He cites the anthropological work of Durkheim and Mauss (1963 [1903]) in
support. But their claim contains no empirical evidence whatsoever. Moreover,
such life-form divisions occur in societies that are not, and likely never were,
organized into moieties. Mayr contends that "the first creatures to be named are,
of course, those of immediate concern to man" (1982:134). Again, ethnobiology
provides scant evidence for, but considerable evidence against, this speculative
proposition.
Origin of the Species and Genus Concepts
197
appear that herbals tend to make their appearance at a rather

advanced stage of city-state development; yet this hardly indicates
"the late advent of interest in plants" (Singer 1927:2) as such.
A common-sense appreciation of phenomenal reality is, it
seems, largely independent of culturally parochial concerns; however, this knowledge of the visible patterns of the organic world
does function provincially, that is, within the local bounds of
everyday experience. Such basic competence appears to be fundamentally the same in all times and places since humankind first
acquired rights to the Linnaean title, Homo sapiens, It would be so
easily accessible to all, and the visible patterns it reveals so
immediately obvious, that there would be no need for members of
a given culture to instruct one another in such knowledge. In a
classic study of Aztec folk botany, for instance, Paso y Trancoso
(1886) presents a much more detailed appreciation of plant life in
general than the Badianus herbal would lead one to suspect. 2
A closer examination of Theophrastus's Historia plantarum,
which provides the model for many later European herbals, shows
that a disucssion of cultural virtues is predicated on the following
premises: that shared virtues are necessary consequences of
similarity in underlying essential nature (physis), and that there is
often no better sign of similarity in nature than a popular intuition
of overall resemblance in readily visible morphological aspect
(eidos, idea, schema, morphe). In other words, learned communication presupposes the principled acceptance of morphological
types, as Aquinas to aptly noted ("figura est signum speciei in
rebus naturalibus" [Summa theologiae III, 74, iii]).
Similar considerations apply to the folkbiology of any society.
There is, it appears, a (universal) presumption to the effect that
visible organic types have underlying natures, and that these
essences provide a principle of natural causality for manifest
organic regularity: "Thus, for Rofaifo [of New Guinea] species
share an essence w h i c h . . , immediately renders the idea, species,
intelligible in a natural (biological) sense" (Dwyer 1976a:433).
Cross-culturally, people presume that living kinds have (possibly
2. Similarly, with respect to the folk classifications of the Tzeltal M a y a of
Mexico and the A g u a r u n a Jivaro of Peru, Berlin notes: " O r g a n i s m s are g r o u p e d
into n a m e d classes primarily on the basis of overall perceptual similarities. This
finding would seem to controvert the view that preliterate m a n n a m e s and
classifies only those organisms in the e n v i r o n m e n t that have s o m e i m m e d i a t e
functional significance for survival. M o r e than one-third of the n a m e d plants in
both Tzeltal and A g u a r u n a , for example, lack any cultural utility, and these are
not pestiferous plants that m u s t be avoided due to poisonous properties or the
like" (1978:10--11).
198
SCOTT ATRAN
unknown) natures with propensities responsible for the readily

perceived regularities of those kinds. So, for example, legless tigers
are still classed with animals considered quadrupeds "by nature"
(cf. Sperber 1975:22). To paraphrase Aristotle in this regard: It is
natural for tigers to be four-legged, that is what we expect to
happen unless something physically hinders their normal maturation" (Physica 199b15). This presumption that a nature and its
propensities produce characteristics typical of a kind -- whenever
normal conditions obtain -- underlies the taxonomic stability of
living-kind terms despite obvious token variation among individuals of a kind (Atran 1987).
There is, then, a pretheoretical sense in which "nature" is distinguished from the artificial, on the one hand, and the supernatural, on the other. In this pretheoretical view, natural things,
like a robin, a rock, or Robert, differ from robots and the
Redeemer by reason of immanent causality; this is, in virtue of
those causal factors that are peculiar to the type of thing and make
it whatever it is -- a bird, a stone, or a man. The difference
between Aristotle's idea of "nature" (physis) and, say, the notion of
"nature" (unuq) entertained by the Bunaq of Timor (Friedberg
1982:1350) is simply this: whereas humans the world over ordinarily presume that each distinct living kind has its proper nature,
Aristotle further assumes that all the distinct natures of folktaxonomic living kinds (as well as of those nonliving sorts modeled
on the living) are somehow causally connected. That is why
Aristotle's physis has the dual meaning of "a given kind" and of
"Nature" in general (Atran 1985b).
After Aristotle and Theophrastus, a general disregard for
careful morphological description and the degenerate practice of
relying on authority rather than on first-hand experience increased
in classical antiquity. To be sure, in Historiarum mundi Pliny
defends experience as "the best teacher," in contrast to the
degenerate practices of the schools, where "it is more agreeable to
sit on b e n c h e s . . , than to go out into deserted places and look for
different herbs at each season of the year" (XXVI, 6). Still, Pliny's
own descriptions are plagued by the very tendencies that he
condemns and are, on the whole, inferior to the descriptions of
Theophrastus. Unlike Theophrastus, or even Pliny, Dioscorides'
verbal descriptions are tightly restricted to materia medica; while
after Krateuas, figures tend to be ever more schematic and bare of
detail (see Singer 1927).
Through the Renaissance, many scholastic "naturalists" merely
assumed that the local flora and fauna of the temperate regions of
Europe could be exhaustively apportioned among the Mediter-
199
ranean plant and animal types depicted in ancient sources. The

experiential foundations of those ancient works, though, had since
ceased to be pertinent. 3 This is because common-sense classifications are relevant merely to a local environment. Only after the
process was reversed by the great German, Dutch, and Italian
herbalists of the sixteenth and seventeenth centuries was classification again grounded in intuitions of natural affinity. These later
herbalists persisted in using the Latin name and description as a
nomenclatural type to which local species could be attached;
however, they no longer simply assumed an ancient type to be
locally represented in the same fashion (or event at all). Thus
3. Judging by the number of extant copies, two of the most important early
botanical treatises were Pseudo-Apuleius's fourth-century Herbarius (Voigts
1978) and Pseudo-Dioscorides' fifth- or sixth-century Ex herbisfemininis (Riddle
1981). Both works draw on Dioscorides but are much more restricted in scope.
The Herbarius portrays some 130 plants, while the Ex herbis depicts seventy-one
herbs. The Herbarius (cf. Singer 1927) and Ex herbis (cf. K~istner 1896) often
simply rely on arguments from authority, with later copies especially containing
figures of generally poor quality and frequent mismatches between ancient and
vernacular names. According to Singer (1927:142), the earliest naturalistic plant
drawings may be found in the twelfth-century Oxford Pseudo-Apuleius manuscript, Bodley 130.
Yet, early on at least, the situation seems to have been somewhat less
disastrous than is generally supposed. One rendition of the Herbarius, for
instance, replaces some of the entries with others that represent plants actually
grown in Central Europe (Landgraf 1928). More significantly, it appears that the
original author of Ex herbis was every bit as circumspect as the great sixteenthcentury herbalists in selecting and modifying Dioscorides' descriptions of
southeast Mediterranean plants to conform to what was likely the author's own
southwest European flora (Riddle 1981). But the originality and accuracy of Ex
herbis should not be exaggerated. There is no evident conceptual advance over
Dioscorides' textual descriptions or Krateuas's illustrations and there is decidedly
less information overall. The texts and drawings frequently do not allow ready
identification of species by a novice; and the contention (Riddle 1981:46--47)
that such entries were "probably" designed for the initiated as generic summaries
of related species is anachronistic. The author of Ex herbis does not represent
overarching morphological relationships in a manner that is uniformly applied to
the entries.
In any event, by the late Middle Ages the state of the art had largely
deteriorated. The chief reference for "Aristotelian" botany in the late Middle
Ages was not Theophrastus, but a bogus version of Aristotle's lost De plantis
composed by Nicolaus of Damascus. This work, which continued to be attributed
to Aristotle into the twentieth century, comprised the core of Roger Bacon's and
Albertus Magnus's thirteenth-century botanical works and did much to vitiate
their occasional empirically well-grounded observations. Even the dominant
botanical treatise of the sixteenth century, Le grand Herbier (or The Grete
Herball) was largely a third-hand mutilation of classical sources, fancifully
illustrated and generally mediated by the second-hand descriptions of Avicenna,
Albertus, and others.
200
SCOTT ATRAN
Fuchs, o n e o f the G e r m a n "fathers of b o t a n y , " r e m a r k s that

Centaurio minore (i.e., Centaurium umbellatum) "grows in wet
and d a m p places, n o t e s D i o s c o r i d e s . . . b u t I h a v e o f t en o b s e r v e d
in o u r G e r m a n y that it g r o w s in hard, dry, grassy fields and
meadows" (1542:386).
A c c o r d i n g l y , the search for d u b i o u s a n ci en t so u r ces o f c o m m o n - s e n s e truths a b o u t o r g a n i s m s in the local e n v i r o n m e n t - - with
the c o n s e q u e n t d i s t o r t i o n of t h o s e truths - - was t r a n s f o r m e d into a
c o n c e r t e d effort to assimilate a n c i e n t and f o r ei g n m a t e r i a l to local
k n o w l e d g e as far as possible. A l t h o u g h this r e n e w e d effort at local
d e s c r i p t i o n p r o c e e d e d in a s o m e w h a t r e f r a c t o r y m a n n e r in o r d e r
to c o r r e s p o n d as closely as possible to the n o m e n c l a t u r e of the
ancients, it is p r e c i s e l y the search f o r such c o r r e s p o n d e n c e that
p r o d d e d the herbalists to fix a m e d i u m o f c o m m u n i c a t i o n and to
establish a s h a r e d r e p o s i t o r y o f d a t a a b o u t the living world. In so
doing, the herbalists m a n a g e d to go b e y o n d c o m m o n sense by
t r a n s c r i b i n g folk u n d e r s t a n d i n g in a m a n n e r that c o u l d b e transm i t t e d across local b o u n d a r i e s of time and place. T h e i d e a of a
w o r l d w i d e system, o r m e t h o d , b e c a m e c o n c e i v a b l e ; but only
b e c a u s e c u s t o m a r y intuitions o f natural affinity, w h i ch ar e o r d i narily restricted to a local e n v i r o n m e n t , w e r e r e c o v e r e d and fixed
in such a way that they c o u l d be p r o v i d e d s t an d ar d s o f e x p r e s s i o n
for c o m p a r i s o n a n d contrast. 4
A principal aim of what follows, then, is to rectify an unfor-
4. Tournefort well understood that ancient knowledge was more thorough

than surviving texts and distorted derivatives might indicate: "The ancients did
not have the aid of engraving for leaving behind the figures of the plants they
used. It was not their custom to make exact descriptions. It seems they counted
more on tradition than their writings; and from this viewpoint they believed it
sufficient to propose the plants most known to them in their time as models for
facilitating knowledge of those that were not. Thus they contented themselves
with such overall comparisons without describing either ]the familiar model or
novel plant] exactly. But things have since changed. That which was so familiar to
them is a mystery today, and in the absence of knowledge as to these first models,
we find only doubt and obscurity in their books" (1694:7--8). Tournefort also
recognized that ancient knowledge was grounded in local understanding and was
not meant to cope with novelty on a worldwide scale: "All of these ]ancient]
names were founded only upon particular viewpoints: one could not foresee that
one day they would have to serve for generic names, that is, names that would be
appropriate for all the species of genera that one would establish in the course of
time. Hence, we cannot complain that the ancients did not reduce this science to
its veritable principles. It was only the experience of many centuries which could
show the rules that one would have to follow in imposing names." But he did not
realize that such local understanding was substantially independent of cultural
functions and not so spotty with respect to the local flora and fauna as herbals
might suggest.
201
tunate but pervasive ignorance of the anthropological foundations

of biological thought that has led to serious misconstruals of the
cognitive foundations and historical course of classificatory knowledge. The notion of "common sense" is used here with systematic
ambiguity to refer to the processes as well as the results of
ordinary, factual thinking. By "common sense" is thus meant that
which is held cross-culturally to be, or to be responsible for,
manifestly perceivable empirical fact. Interpreted in this way,
common sense also includes those statements that pertain to what
is likely an innately grounded, and species-specific, apprehension
of the spatio-temporal, chromatic, chemical, and organic world in
which we, and all other human beings, live our usual lives. To put
it simply, owing to our cerebral architecture we are all constitutionally disposed to believe that the world of everyday experience
is composed of natural chemical and biological kinds whose
exemplars manifest definite colors, change in time, and are locally
distinguished by their relations in space.
Regarding human beings' appreciations of living kinds in particular, two decades of intensive empirical and theoretical work in
ethnobiology seem to reveal that folkbiological classification is
taxonomic, being composed of a rigid hierarchy of inclusive
classes of organisms, or taxa. At each level of the hierarchy, the
taxa, which are mutually exclusive, exhaustively partition the
locally perceived flora and fauna. Lay taxonomy, it appears, is
universally and primarily composed of three absolutely distinct
hierarchical levels, or ranks: the levels of unique beginner, basic
taxa, and life-forms (see Berlin, Breedlove, and Raven 1973).
The "unique beginner" refers to the ontological category of
plants or animals (see Keil 1979). Some cultures employ a special
marker for the unique beginner, like the numerical classifier tehk
for plants, as used by the Tzeltal Maya (Berlin, Breedlove, and
Raven 1974). Others use a descriptive phrase, such as "the hairs of
the earth" (muk gobul nor) of the Bunaq of Timor (Friedberg
1972). Yet others have no word or ready-made phrase for plant or
animal, although from an early age all humans seem to conceptually distinguish these cateogories, as indicated by studies of
young Mayan (Stross 1973) and American children (Dougherty
1979; Macnamara 1982), New Guinea highlanders (Hays 1983),
Indonesian natives (Taylor 1984), and so forth.
The "basic" level is logically subordinate, but psychologically
prior, to the life-form level. Ideally it is constituted as a fundamentum relationis, that is, an exhaustive and mutually exclusive partitioning of the local flora and fauna into well-bounded
morpho-behavioral gestalts. For the most part, taxa at this level
202
s c o t t ATRAN
correspond, within predictable limits, to those species of the field

biologist that are spatially sympatric (i.e., coexisting in the same
locality) and temporally nondimensional (i.e., perceived over at
most a few generations), at least for organisms that are manifest,
including most vertebrates and flowering plants (cf. Mayr
1969:37). Because the frontiers of a cultural group do not always
correspond to the boundaries of a set of sympatric species,
partitioning can fall short of the ideal: for example, migrating birds
may be only intermittently or vaguely represented.
But this basic folk kind also largely conforms to the modern
genus, being immediately recognizable both ecologically and morphologically. As we shall see, this state of affairs underlies much
of the confused and seemingly inexhaustible controversy over
whether the species (cf. Diamond 1966) or the genus (cf. Bartlett
1940) constitutes the psychologically and historically primitive
grouping. In this regard, some ethnobiologists refer to basic taxa as
"generics" (Berlin 1972), while others term them "speciemes"
(Bulmer 1970). The controversy is, in fact, a dead issue. This is
because the species-genus distinction is not pertinent to a local
understanding of the natural environment, that is, within phenomenal dimensions of everyday human experience. That is why I
have designated the basic folkbiological kind "generic-specieme."
Historically, a taxonomic notion of the species as a perpetually
self-reproducing unity from a common seed was introduced in
Cesalpino's original treatise on systematic botany (1583:1, 28),
and the conception of the gvnus as a perceptually and mnemonically privileged rank immediately superordinate to the species was
first codified by Tournefort (1694:13--14). But these historical
occurrences pertain to the necessity of constructing a worldwide
classificatory system, and are not motivated by local concerns.
The "life-form" level further assembles the generic-speciemes
into larger exclusive groups (e.g., tree, bush, grass, moss, mammal
or quadruped, bird, fish, bug, etc.). A salient characteristic of
folkbiological life-forms is that they seem to partition plant and
animal categories into contrastive lexical fields. The system of
lexical markings thus constitutes a pretheoretical fundamentum
divisionis into features that are positive and opposed. The opposition may be along a single perceptible dimension (size, stem
habit, mode of locomotion, skin covering, etc.) or simultaneously
along several dimensions (cf. Brown 1977, 1979). By and large,
plant life-forms do not correspond to scientific taxa, whereas
animal life-forms more or less conform to modern classes, excepting the phenomenally "residual" categories of "bugs," "worms" and
"insects."
203
Such uniform taxonomic knowledge, under such diverse sociocultural learning conditions, likely results from certain regular
and domain-specific processes of human cognition, though local
circumstances undoubtedly trigger and condition the stable forms
of knowledge that are attained. Species recognition is perhaps a
trait of nonhuman species as well. Yet, plausibly, only humans
define species by presuming them to have underlying natures, or
essences, that underpin the taxonomic stability of phenomenal
organic types despite obvious variation among individual exemplars. Moreover, humans apply a recursive procedure for further
sorting species into higher-order categories. 5 Without pencil and
5. The claim for universal principles of folkbiological taxonomy is not for the
universal status of particular taxa, only for taxonomic categories. Taxa are
particular groups of organisms (e.g., dogs, trees). Categories are ranked classes
of taxa (e.g., generic-specieme, life-form). Taxa and categories thus comprise
different logical types. The categories of generic-specieme and life-form are
universal. The delimitation and placement of particular taxa are not. Applied to a
local biota, universal taxonomic principles (including presumptions of underlying
natures) tend to yield basic-level groupings that correspond to biological species,
at least for the phenomenally salient vertebrates and flowering plants.
Formal taxonomic constraints are deductive and inductive. The deductive
constraint requires transitive inference as to group adherence: if one discovers a
new kind of oak, then one knows it to be a tree. The inductive constraint allows
for inferences as to the general distribution of taxonomic (and ancillary morphoecological) features throughout the local flora and fauna: if one discovers two
organisms to possess a feature, then one may infer that the feature likely belongs
to all organisms in the lowest-ranked taxon containing the two. The inductive
character of life-forms pertains primarily to ecological and morphological relationships between species. Some cultures classify bats with birds, other place bats
with quadrupeds, still others accord bats their own (monogeneric) life-form
status, depending upon the bat's perceived relationships with the totality of the
local fauna (and flora). As the distribution of ecological boundaries and morphological characters varies from one locale to another, so may life-form boundaries.
Whatever the case, universal taxonomic principles operate just the same.
Given these principles, then, to class an organism under a life-form is not
simply to presume that it has the nature of that life-form; rather, it is to predicate
for the organism membership in one or another of the generic-speciemes that has
as part of its nature the nature of that life-form. So, as Theophrastus (Historia
plantarm I, iii, 2) stresses, when mallow, which is normally not like a tree, grows
tall like a tree it departs from its "essential nature" (physis). In this case mallow is
said to be merely "tree-like" (apodendroumeni), and not a tree "by nature"
(physei). A n d although one often hears native Indonesians - - or folk anywhere,
for that matter - - saying of a particular small sapling that "this herb is a tree," of
the same small sapling they will insist that "this is not a (member of the)
h e r b a c e o u s . . , class, it is a tree" (Taylor 1978--1979:224).
Such absolute and essential ranking of living kinds is apparently unique to that
cognitive domain. The structure of artifact groups, which is often confounded
with that of living kinds, is in fact quite different. For one things, that taxa of the
204
SCOTT ATRAN
p a p e r , h i e r a r c h i c a l sorting p r o c e d u r e s a p p e a r to b e severely l i m i t e d
b y c o n s t r a i n t s o f m e m o r y (cf. R a v e n , Berlin, a n d B r e e d l o v e 1971).
But the i m p l i c a t i o n s of this r a n k i n g p r o c e s s for i n d u c i n g the causal
relationships b e t w e e n taxa a r e p r o f o u n d l y e x t e n d a b l e . Historically,
g r e a t e r q u a n t i t y o f i n f o r m a t i o n as to n u m b e r a n d k i n d s o f o r g a n isms, a n d g r e a t e r quality of i n f o r m a t i o n as to their r e l a t i o n s h i p s ,
c o u l d b e p r o c e s s e d b y i n t r o d u c i n g a d d i t i o n a l r a n k s into a w o r l d w i d e t a x o n o m y . T h e d e d u c t i v e f r a m e w o r k of this rigid transitive
h i e r a r c h y w o u l d d o u b l e as an i n d u c t i v e c o m p e n d i u m o f life in
general; for e x a m p l e , given a p r o p e r t y that is o b s e r v e d in two
o r g a n i s m s , o n e might p l a u s i b l y s u p p o s e that all o r g a n i s m s b e l o n g ing to the l o w e s t - r a n k e d t a x o n c o n t a i n i n g the two o u g h t to h a v e
the p r o p e r t y .
T h e folk c a t e g o r i e s o f g e n e r i c - s p e c i e m e a n d life-form, which
r a n k " g r o u p s within g r o u p s , " thus c o n s t i t u t e the f u n d a m e n t a l
setting for h u m a n d k i n d ' s o r d i n a r y a p p r e h e n s i o n o f the local flora
a n d fauna, b u t p r o v e to b e i n a d e q u a t e for c o m p r e h e n d i n g the
living w o r l d at large. N e v e r t h e l e s s , t h e s e u n i f o r m i t i e s have historically c o n d i t i o n e d n a t u r e ' s t r a n s i t i o n f r o m setting to subject. T h i s is
e v i d e n t in the s y s t e m a t i c e l a b o r a t i o n o f the t a x o n o m i c distinction
b e t w e e n s p e c i e s a n d genus.
T h e p r i m a r y o p e r a t i o n in the classification o f living kinds,

w h e t h e r for the field naturalist o r the m u s e u m t a x o n o m i s t , is the
a g g r e g a t i o n o f o r g a n i s m s into basic-level taxa. This m e a n s partitioning a set o f i n d i v i d u a l o r g a n i s m s into classes, o r taxa, that a r e
in s o m e sense h o m o g e n e o u s with r e s p e c t to specified attributes:
"Such a p r o b l e m faces us w h e n w e a t t e m p t to sort i n d i v i d u a l
same category are disjoint precludes artifact groupings from entering into ranked
taxonomies. Not only can artifacts belong to more than one contrasting "taxon"
within an inclusion series (a wheelchair as both "furniture" and "vehicle", a piano
as both "musical instrument" and '~furniture"), but a given item may belong to
different series (the same item as a crate used for packing furniture, or as a table
used as a piece of furniture). Artifacts have no intrinsic essences, but are defined
by the functions they are meant to serve (a given surface may be considered a
table top or a seat, depending on the context; a wastepaper basket can literally
become a stool if habitually oriented accordingly). Further, artifacts fail to meet
the deductive and inductive requirements of ranked taxonomies: "car-seats" may
be judged varieties of "chair", but not of "furniture," even though "chair" is
normally thought of as a type of "furniture"; and it is hardly plausible that we
induce, say, that tables are quadrupedal from the fact that tables are normally
observed to have four legs (see Atran 1987).
205
organisms into 'species"' (Jardine 1969:37). This basic partitioning

of the living world into homogeneous groupings involves the
construction of configurally well-bounded "morphotypes." To do
this, the taxonomist begins by looking over the material,
noting (subconsciously) striking features of the variation, including similarities and differences. Then he will being to sort
specimens into piles, each containing specimens that, to his eye,
show morphological continuity and homogeneity. This process
will c o n t i n u e . . , until he is satisfied that he can make no further
improvements . . . . What should be pointed out here is that all
this is done without any [discursive] concept of character being
used at all; what the taxonomist is doing is comparing mental
images of whole specimens as wholes; he is comparing what
have been called Gestalten. (Cullen 1968:176)
By comparing whole images of specimens, "observational homologies" are established between them (Inglis 1966), that is, a
judgment of absolute resemblance is made between the corresponding structures of different exemplars of the same type. The
corresponding homologous structures of the various exemplars of
the type are deemed wholly equivalent for classificatory purposes,
and logically identical with respect to the ideal-typical pattern, or
"species" structure (Regan 1926). The various exemplars thus
become token "namesakes" of the species-type (Owen 1866). This,
in essence, is the Aristotelian notion of the basic-level organic
grouping, or atomon eidos:
With regard to animals, there are those which have all their
parts mutually identical. . . . Some parts are specifically identical
in form, for example, one man's nose and eye are identical with
another's nose and eye; one's flesh is identical with another's
flesh, one's bone with another's; and the same applies to the
parts of a horse, and of such other animals as we consider
identical to one another by species (eidei); for as the whole is to
the whole, so every part is to every part. (Aristotle, Historia
animalium 486a 14 f.)
Yet, as Simpson notes, in the practice of classification the
abstraction of the biological species "is not always the first operation, and in identification it never is" (1961:23). In many cases, it
seems that generic, rather than specific, groupings are easiest to
sort. This is because the exact delimitation of biological species
as breeding isolates is often difficult in practice owing to the
206
SCOTT ATRAN
complication of perceptually aberrant mutants, incomplete fossil

records for historical or paleontological species, and phenotypically similar "sibling" species that can usually be identified only by
careful observation and experiment over time.
In Cain's view, the genus is "the smallest 'kind' or 'sort' recognisable without expert study" (1956:108). For plants, observes
Cronquist, "if the circumstances permit we try to define genera in
such a way that one can recognize a genus from its aspect, without
recourse to technical characters not readily visible to the naked
eye" (1968:30) Sprague describes the basic sorting process so:
A taxonomic botanist beginning the study of . . . a genus, will
set out all his herbarium and other specimens together with
the best available illustrations so that he may obtain a bird's eye
view of the whole . . . . In such preliminary work the faculty of
appreciating what is called the "facies" of the plant is invaluable.
. . It is a general effect produced on the eye by the sum total of
all the visible external characters, many of which are not
actually employed in technical descriptions owing to the difficulty of expressing them in precise terms. The increase of
precision in modern technical d e s c r i p t i o n s . . , has been accompained by a certain deterioration in the art of plant description
The better sixteenth-century descriptions of plants, such as
those of Valerius Cordus, can be best understood by treating
them as word-pictures designed to supply mental images.
(Sprague 1940:449).
Similarly, in zoology "the basic or one might say primitive step

in classification" involves "nontechnical recognition," which "is
normally not by separate characters but by a mental image of the
whole animal" (Simpson 1961:12). Thus, "we can visualize that
primitive man, in the process of domestication, picked plants [and
presumably animals] on the basis of their generic features without
discrimination as to species" (Li 1974:723).
This historically and conceptually primary stage of scientific
classification, then, would appear to correspond to the basic
operation in folkbiological classification the world over. According
to Bartlett, the concept of the genus is the most natural and useful
level of classification, inasmuch as it is "the smallest group that
everyone might be expected to have the name for in his vocabulary" (1940:351) Berlin notes that basic-level folk taxa fit this
requirement because they are, in fact, "the first to be encoded in
the ethnobotanical lexicons of all languages" (1972:55). The same
207
appears to be true of folkzoological taxonomies (Berlin, Breedlove

and Raven 1973).
The principal characteristic of the basic "folkgeneric," as Berlin
(1978) calls it, is its readily perceived "aspect." Hunn (1975a)
reports that the "gestalt" character of the folkgeneric lends itself to
"instantaneous recognition"; but he further argues for an intuitive
justification of folkgenerics on empirical grounds of "inductive
recognition" of a "multiplicity of [visible[ distinctions." Within any
given local environment it is usually the case that no two coexisting
folkgenerics occupy the same habitat, so that "it is inevitable that
differences [between basic folk taxa] . . . will be reflected morphologically and/or behaviorally" (Hunn 1975b:320; 1976:523n).
In a study of seals and their allies (Pinnipedia), Scheffer (1958)
indicates some further properties of such "genera" usually distinguished by the natives: (a) at least one variate does not overlap
with other genera when comparing individuals for same age and
sex; (b) the way of life is distinctive; (c) the breeding range is
confined to one or two broad niches; and (d) different genera
do not interbreed in the wild. Although some taxonomists tend
to view modern genera similarly (cf. Inger 1958:383), Simpson
objects that these features "are all characteristic of . . . the evolutionary species, not genus"; as for the native, such an account "has
some validity for the species, as field workers know," but not for
the genus (1961:189). Indeed, for Bulmer and Tyler (1968:350),
basic folk taxa, or "speciemes," are "logically comparable to
modern species." 6
Now, if Scheffer's point is that the natives first tend to class the
Pinnipedia into groups possessing the features listed above, how
can such properties attach to species, since, according to Simpson,
the species is supposedly rarely the first step in "nontechnical,"
or "primitive" classification? And if, as Berlin (1973:267--268)
concedes, basic folk taxa closely correspond to scientific species
"in Bulmer's sense," that is, in regard to "multiple distinctions of
appearance, habitat and behavior," how can they also resemble
modern genera? The apparent dilemma is resolved once it is
realized that the distinction between species and genus is largely
irrelevant to a basic appreciation of the flora and fauna of a local
environment; historically, the distinction only assumes a modern
character in connection with Europe's Age of Exploration, that
6. For rather obvious reasons zoologists and ethnozoologists tend to
highlight basic groupings in terms of behavioral and specific features, whereas
botanists and ethnobotanists are apt to stress morphological and generic
characters.
208
SCOTT ATRAN
is, in the context of a search of worldwide order -- an order

transcending local concerns.
Within any local community the layman readily perceives
"gaps" between groups of organisms. For the most part, these
apparent discontinuities in the local flora and fauna correspond to
species differences. Among the Tzeltal Maya, for example, Hunn
(1976) reports that nearly all (95%) animal generics correspond to
recognized scientific taxa, three-fourths (75%) to scientific species
and more than half (57%) to "isolated species" having no congeners in the local area. So, in the majority of cases, species and
genus are extensionally equivalent and hence cannot be distinguished perceptually.
When congeneric species are perceived as each having generic
status it is usually because they are ecological isolates. As Dwyer
notes in regard to the Rofaifo of the New Guinea: "when closely
related species co-occur, they will diverge significantly in the
ecological strategies they adopt and often in the morphological
correlates of those strategies" (1976a:430; cf. Diamond 1972).
This is espeically so if the species fall within families that are
locally monogeneric or minimally polytypic. That such cases are
not infrequent may be expected, given the general disposition of a
local flora and fauna. In the case of the local flora of the Bunaq
of Timor, "one may note that . . . certain botanical genera are
represented by few species, certain families by few genera and that
the Bunaq therefore have but a partial vision of the botanical
world wherein certain plants seem to them completely isolated and
hence determinable without confusion" (Friedberg 1970:1128).
Thus is happens that folk nomenclature contains numerous basic
terms that denote only isolated species.
For those instances where generics seem to be underdifferentiated, that is, where a folkgeneric subsumes two or more species,
it is often legitimate to ask, along with Bulmer: "How many of
these cases include species which the professional botanist can
only distinguish by very fine points so that the layman could
justifiably regard them as natural units sharing a wide range of
attributes? How many of them include only plants [or animals]
which are relatively unfamiliar?" (1970:1088). Indeed, because the
genera of a local family usually occupy a single adaptive zone in a
particular community, identifying the congeneric species of a
polygeneric family can pose problems of recognition even for the
trained taxonmist because of overlap in aspect and ecological
proclivity (cf. Diver 1940).
In sum, within any given local community a principled distinction between genus and species is usually impossible perceptually
209
because most genera are monospecific. The distinction is frequently also irrelevant conceptually inasmuch as the species of
locally monogeneric or minimally polytypic families may have the
same sorts of morphological, geographical, and behavioral correlates as the genera of locally polygeneric families. In such cases,
the perceptible morpho-ecological distance between species of the
monogeneric or minimally polytypic family m o r e or less corresponds to that between genera of the polygeneric family (cf. Berlin
1982a). Finally, the distinction is often inconsequential to anyone
but a geneticist or microevolutionary theorist (e.g., sibling species
that occupy the same niche and manifest virtually identical phenotypes). For the most part, then, local species are marked by
"generic gaps" in the local e c o n o m y of nature, and have no rivals
to compete with for generic status. 7
Most of the time, this generic gap is filled by what Mayr has
called "the nondimensional species": "At a given locality a species
of animal is usually separated from other sympatric species by a
complete gap. This is the species of the local naturalist, the species
of Ray and Linnaeus. It may also be called the nondimensional
species because it lacks the [evolutionary] dimensions of time and
space. Combining properties of a single local population, the
nondimensional species can usually be delimited unequivocally"
(1969:37). Mayr misleads, however, in implying that the manner
of delimiting basic local kinds is the same for the folk naturalist as
it was for Ray and Linnaeus, namely, as a breeding unit: '~The
word species in biology is a relational term: A is a species in
relation to B because it is reproductively isolated from them. It has
its primary significance with respect to sympatric and synchronic
populations" (Mayr 1969:26). F o r folk, as for Aristotle and later
E u r o p e a n herbalists, reproductive criteria, though usually suf7. As Darwin notes: "We often take, I think, an erroneous view of the
probability of closely-allied species invading each other's territory when put into
free communication. Undoubtedly, if one species has any advantage over
another, it will in a very brief time wholly or in part supplant it; but if both a r e
equally well fitted for their own places, both will probably hold their separate
places for almost any length of time. Being familiar with the fact that many
species, naturalised through man's agency, have spread with astounding rapidity
over wide areas, we are apt to infer that most species would thus spread; but...
the species which become naturalised in new countries are not generally closely
allied to the aboriginal inhabitants, but are very distinct forms, belonging in a
large proportion of cases, as shown by Alph. de Candolle, to distinct genera. In
the Galapagos Archipelago, many even of the birds, though so well adapted for
flying from island to island, differ on different islands; thus there are three
closely-allied species of mocking-thrush, each confined to its own island"
(1883:356).
210
SCOTT ATRAN
ficient, are not always necessary. T h e necessary condition is an
ecological distinction that marks a m o r p h o - g e o g r a p h i c a l gap in the

readily perceptible local e c o n o m y of n a t u r e /
Necessary breeding criteria were initially introduced into classical natural history in o r d e r to fix species status w i t h i n a particular c o m m u n i t y , so that a systematic c o m p a r i s o n and grouping of
species could be m a d e across communities. T h e idea was to provide a universal (or "natural") physical means (or "mechanism")
for species production, that is, for the a p p e a r a n c e and maintenance of morphological regularity in any given location. It was to
be wholly internal to the species - - part of its "essence" - - and
i n d e p e n d e n t of external factors c o n n e c t e d with environmental
regulation. A p e r m a n e n t fixation of the limits of m o r p h o l o g i c a l
variation (the degree of permissible morphological difference) for
the species would justify abstracting a m o r p h o t y p e by omitting all
but its most usual features as irrelevant. With the m o r p h o t y p e thus
extracted f r o m context, the c o m p a r i s o n and grouping of species
could be m a d e entirely on the basis of cross-specific regularities in
m o r p h o t y p i c a l characters. Because universal "natural" t a x o n o m y
was w o r k e d out by botanists who, following Aristotle, first thought
that most, if not all, plants propagate asexually, criteria of genealogical linkage rather than cross-fertility were initially chosen to
ensure conspecificity.
II
Early E u r o p e a n classifiers m o r e or less followed the habit of
local folk in reducing unfamiliar fauna and flora to locally familiar
8. In cases of species recognition for asexual organisms, Mayr acknowledges

that the only "objective" criteria consist "in the fact that each of the morphological entities, separated by a gap from similar entities, seems to occupy an
ecological niche of its own" (1969:31). Moreover, it seems the isolating mechanisms that prevent the interbreeding of sympatric populations are invariably
associated with niche specialization. Because such specialization also characterizes uniparentally reproducing organisms, there seems to be prima facie
justification for the claim that the fundamental "taxonomic species" (Blackwelder
1967) is a "morphological-geographical species" (Davis and Heywood 1963),
that is, an "ecological species" (MacArthur 1972). Perhaps becausC botanists
are more readily confronted with asexual organisms, they tend to prefer an
"ecological" to a "reproductive" species. Yet, the two notions are inextricably
bound to one another. Even uniparentaUy reproducing organisms come to
occupy their position in the economy of nature partly in virtue of their own
genealogical heritage and partly in virtue of the place afforded them by other
species that are characterized by reproductive isolation.
211
sorts of plants and animals. Friedberg describes the process for the
Bunaq of Timor:
Each time we asked them to name and classify a rare p l a n t . . .
they would try to attach it to . . . a series [i.e., a basic-level
grouping], always on the basis of the plant's appearance -- an
appearance which is as much its morphology, anatomy as its
odor or the texture of its fibers. On thus has the impression that
if the informant does not know the name of a plant he can
always find one in conformity to the logic of the classificatory
system, by giving it the base name of the series which it most
closely resembles. (Friedberg 1970:1122--23)
Similarly, Dughi notes:
The nomenclature of the ancient Greeks and later that of the
Latins, which we find until the seventeenth century, is not
basically different from popular botanical nomenclature, for
example Provenqal [southern France] nomenclature. Here
drawn around certain paragon plants is a whole tribe of plants
which are sometimes related, sometimes not, but which share
traits of external resemblance with the paragon. (Dughi
1957:136)
Theophrastus appears to treat many of the exotic plants sent back
to Greece from Alexander's expeditions in just this way (Bretzl
1903). 9 The banyan (Ficus benghalensis), for instance, is simply
labeled an Indian "variety" (syka indika) of the common Mediterranean fig (syka, i.e., Ficus carica): "the whole tree is round and
exceedingly large . . . the fruit is very small, only as large as a
chickpea, and it resembles a fig. And that is why the Greeks
named this a 'fig tree'" (Historia plantarum IV, iv, 7).
Now, this strategy, which is based on likeness of the whole or a
part of the habitus, occasionally leads to an appreciation of natural
affinity. For example, the English colonists were thus able to
assimilate the widely divergent species of North American oaks to
the rather narrow range of English trees denoted by "oak": for
example, "shingle oak" (Quercus imbricaria), "scarlet oak" (Q.
coccinea), "dyer's oak" (Q. tinctoria), etcetera. If, however, the
9. Regarding plant classification in classical Chinese encyclopedias, Morton
notes: "A new plant was named by using the classical name of the plant it
resembled, combined with a qualifying word: leading to an almost binomial
nomenclature analogous to that developed by Theophrastus" (1981:60; Needham
1986).
212
s c o t t ATRAN
foreign species is truly exotic, such a strategy inevitably fails. Thus,

it seems that when wheat and sorghum were first introduced to the
Tzeltal Maya they were referred to as "Castillian maize" (kaslan
eisim) and "Moor's maize" (m6ro eisim), respectively. Initially,
then, these exotic forms appear to have been considered folkspecific variations of the indigenous generic-specieme eisim, that
is, maize (Berlin 1972). Conversely, the English colonists likely
thought of maize as an indigenous variety of wheat when they first
called it "Indian corn."
Once an unfamiliar plant or animal is labeled and identified by
perceptual analogy with the closest indigenous facies, though, it
may itself assume a marked role in the local economy of nature.
Accordingly, increased familiarity with the once unfamiliar plant
or animal may eventually lead to the structuring of a distinct facies.
When this happens the once unfamiliar species usually drops the
base name of the indigenous generic-specieme to which it was
originally attached and assumes its own unique label. Thus, for the
folk of France today, the Zea plant is more commonly referred to
as ma~'sthan as the bl~ de turquie of former times. 1
The same cognitive process that governs the incorporation of
exotic species into the local scheme can also regulate the upgrading of native species in conformity with the ever-changing
relationships of local species to one another and to local folk.
Again for the Tzeltal, the "armadillo-eared oak" (ciknib hihte ~)
may once have been viewed as but one of a number of specific
variations of the generic-specieme "oak" (hihte ?) along with such
other folkspecifics as "white-footed oak" (sakyok hihte ~) and
"custard-apple oak" (k'ewes hihte?). At present, the armadilloeared oak" is separated from these other oaks. The latter group of
broad-leafed oaks are now included under the optional rubric
"true oak" (bac'il hihte?), or merely "oak" (hihte?), whereas
the "armadillo-eared oak" is referred to these days as simply
"armadillo-eared" (ciknib), with the appellation "oak" (hihte ?)
being entirely optional. Ciknib has thus attained the grade of a
generic-specieme that encompasses the various native small-leafed
oaks, while hihte ? is left with only the broad-leafed oaks as
specifics.
On occasion, though, the binomial qualification may linger; for
example, while "Castillian maize" and "Moor's maize" currently
rank as generic-speciemes, they maintain binomial form. What
distinguishes today's situation from the past is that the indigenous
generic-specieme is nowadays optionally marked by the attributive
10. Turkeywas often taken by Europeans as a proximate source of exotica.
213
"true" (bac'il); hence, it too is nominally binomial in structure

(bac'il ?isim): "Time and usage, however, will tend to neturalize
the marking properties of the attributive forms kaslan and m6ro
and the expressions will come to be conceived of as single,
semantic units" (Berlin 1972:75). 11
A similar process is evidently at work in the herbals of the three
"German fathers of botany": Otto Brunfels, Jerome Bock (or
Hieronymus Tragus), and Leonhart Fuchs. lz The main difference
between these Renaissance herbalists and ordinary folk concerns
the original referential type. For the German herbalists, the type is
frequently an ancient rather than a local generic-specieme. Consequently, the process for incorporating new species seems to
function in reverse. Instead of attempting to match foreign species
to local types, it is more often a matter of matching local species to
foreign types, although the principle is basically the same.
Take, for example, the Latin edition of Brunfels's Herbarum
vivae eicones (1530--1536). Here the author refers to the mallow
of the ancients as simply Malva, but labels a second species found
in Germany Malva Equina, or "horse mallow." Yet, in the German
edition it appears that Malva Equina is no longer simply a variety
of Malva because both species are now labeled with binomials:
"horse mallow" is called Rossbappelen, and the original Malva is
termed Gaensbappelen, or "goose mallow." The local generic-
11. This is one reason why folkgenerics cannot be invariably associated with
single lexemes. However, this does not mean that binomials are often associated
with speciemes that fall under recognized supraspecific groupings corresponding
to incipient biological genera. As Dwyer notes, the agreement of binomially
labeled folk groupings with morpho-geographical species is, on the whole,
"exceedingly poor"; this seems to indicate that "binomial species are of more
recent origin in Rofaifo thought than are species designated by mononomials"
(1976a:434). To the extent that binomial groupings are recognized as speciemes
[i.e., as folkgenerics rather than folkspecifics], they may be expected eventually to
acquire a mononomial title. But even if the vestigial binomial label stays (cf.
Strathern 1969), the conceptual status of the grouping will have changed. This
allows that an appreciation of overarching morphological affinity between the old
and new speciemes may persist on much the same footing as other, usually
"covert" (unlabeled), associations (see Atran 1983); ~isim, for example, may be
taken to refer overtly to indigenous maize, but occasionally it may be taken as a
gloss for something like "grains" as well. Still, there is no indication whatever that
folk have another morphological rank in the offing -- a conceptually pinneddown perceptual layer of being over and above that of the specieme or folkgeneric -- that more or less corresponds to the modern genus. In general, lexical
status is only a very rough indication of conceptual status.
12. Despite the somewhat overenthusiastic judgment implied in Sprengel's
(1817-- 1818) attribution of paternity, the motto has stuck.
214
SCOTT ATRAN
specieme has apparently achieved conceptual equality with the

ancient type (cf. Bartlett 1940).
Consider, now, a slightly different example. In the Kreiitter
Buch (1539), illustrated and published in Latin as De stirpium
(1552), Bock distinguishes domestic and wild varieties of the
ancient Melissa (Miitterkraut) from the c o m m o n Melissa vulgaris
(Gemein Miitterkraut). In De historia stirpium (1542), Fuchs
proceeds to give the ancient type the optional marking "true"
(Mellissophylon verum), while the common specieme is accorded
the optional epithet '~false" or "bastard" (Mellissophylon vulgaris
vel adulterinus) -- for, "according to the opinion of popular
herborists there are two sorts of Melissa."
Still, it would be a mistake to conclude that either the binomial
nomenclature of folk or that of the German herbalists evinces the
modern idea of the genus. For folk, binomial sorts generally
occur at a taxonomic level below that of the generic-specieme.
They are, more often than not, culturally idiosyncratic varieties
that do not exhaust the local flora or fauna. Such folkspecific or
folkvarietal taxa, as ethnobiologists call them, usually occur in sets
of two or three members that contrast with one another on the
basis of very few (and frequently only one culturally important)
morphological characters (cf. Berlin 1978). Similarly, for lay
colonists:
In the place of the one European kind of walnut, the Virginia
forests p r e s e n t e d . . , at least a half-dozen, each strikingly unlike
the Old World type . . . . If one type of these peculiarly American walnuts bears today the name of White Walnut [Juglans
alba], it is undoubtedly because the first settlers o~ Virginia,
taking it for a probable equivalent of the English Walnut for
lumbering purposes, found its wood to be by comparison much
lighter in color and named the tree, after the usage of lumbermen, by the color of its wood. The Black Walnut [Juglans nigra]
in like manner obtained its name from the almost blackish hue
of its wood compared with that of the tree of Europe [Juglans
regia]. (Greene 1983, 1:106).
True, for folk, as for the German herbalists, morphologically
similar generic-speciemes are occasionally related binomially. But
this does not mean that binomially labeled generic-speciemes that
share a c o m m o n base name are consistently deemed to group
together at a distinct level of reality, or rank, superordinate to the
basic level of folktaxonomy (as Greene suggests; also Dughi
1957:138; Bartlett 1940:355). In other words, no definite relation-
Origin of the Species and G e n u s C o n c e p t s
215
ship of genus to species can be established between Bappelen and

all the other (disjoint) basic-level groupings n o t e d by Brunfels, any
m o r e than between ? isim and all other Tzeltal generic-speciemes.
This is not to d e n y that linguistically related generic-speciemes
may share recognized m o r p h o l o g i c a l characteristics. Only, such
overarching groupings of affinities o c c u p y neither a u n i f o r m perceptual stratum n o r a fixed conceptual position in 'the classificatory scheme.
W h a t is rightly intimated by G r e e n e and Bartlett is that the
w o r k of G e r m a n herbalists insinuates a shift towards local realism
and away f r o m the antiquated and d e b a s e d practices o f the
scholastic herbalists who were mostly c o n c e r n e d with attaching
local plants to the preexisting names and descriptions of ancient
types (cf. M e y e r 1 8 5 4 - - 1 8 5 7 ; Sachs 1890[1875]. 13 Because plants
of the m o r e temperate regions of E u r o p e were often quite distinct
f r o m the plants described by ancient authors of the M e d i t e r r a n e a n
world, such a literary exercise could only succeed in c o n f o u n d i n g
the true affinities of plants. T h e G e r m a n herbalists of the sixteenth
and seventeenth centuries, like their French, Dutch, and Italian
c o n t e m p o r a r i e s , ~4 often persisted in using the Latin (or Latinized
Greek) n a m e and description as a nomenclatural type to which
local species could be attached; however, unlike their scholastic
predecessors, they did not simply assume the ancient type to be
locally represented.
C o n c e r n with local conditions is particularly notable in Fuchs.
Thus, u n d e r the section tempus in the chapter on Cucurbita Sativa
(i.e., C. lagenaria L.), he notes: 15 "In G e r m a n y the fruit c o m e s later
owing to the chilliness of the country, and yet before a u t u m n is
13. The bookish practices of the schools, which subordinated fieldwork to
literary tradition, were already current in late antiquity despite pleas made by
such as Pliny and Galen in favor of first-hand experience.
14. The outstanding names of the period are centered around two schools:
Ghini's in northern Italy (Bologna and Pisa), which is associated with Aldrovandi,
Turner, Mattioli, Cordus, Gesner, and Cesalpino; and Rondelet's in Montpellier,
which also sustained contact with the previous group through the likes of Clusius,
Lobelius, and Daleschamps.
15. To each plant type, Fuchs allots a chapter divided into the following
sections: (1) nomina provides Greek, Latin, and German synonyms; (2) genera
occurs only with polytypic forms and gives the main variants; (3) forma offers a
brief description of all (or, more often, some) of the mains parts of plants, that is,
root, stem or trunk, flower, leaf, fruit, seed, and branch; (4) locus depicts the local
habitat; (5) tempus describes the schedule and pattern of maturation and
flowering; (6) temperamentum characterizes the subjective "temperature" as
determined by the taste and texture of the plant parts; and (7) vires portrays the
virtues of the plant as rendered by Dioscorides, Galen, Pliny, etc.
216
SCOTT ATRAN
d o n e t h e r e is n o n e left to find." F u c h s c o n s i d e r a b l y a d v a n c e d
the art o f naturalistic illustration a n d d e s c r i p t i o n , a n d d i d n o t
hesitate to c h a l l e n g e the ancients. L a c k i n g a n o t i o n o f c o m p a r a t i v e
m o r p h o l o g y , though, he is d i s p o s e d to e m p h a s i z e only t h o s e
a s p e c t s that s e e m to h i m m o s t c h a r a c t e r i s t i c o f the p l a n t in
question. F o r e x a m p l e , c o n s i d e r a t i o n s of the f o r m a n d figure o f
the fruit l e a d him to p o s i t t h r e e "genera" of C. lagenaria, w h e r e
Pliny h a d m e n t i o n e d o n l y two: "Pliny, in the fifteenth c h a p t e r o f
the n i n e t e e n t h b o o k , m a k e s two cultivated species . . . we, having
m o r e r e g a r d f o r the f o r m a n d figure o f the fruit, h a v e s e p a r a t e d
three." 16
In the a b b r e v i a t e d G e r m a n e d i t i o n o f his work, N e w Kreiiterb u c h (1543), F u c h s a d m i t s only 2 8 9 of the 6 0 0 o r so s p e c i e s
k n o w n to the ancients. T h e others, he argues, a r e n o t p e r t i n e n t to
the e v e r y d a y k n o w l e d g e of the c o m m o n man. T h e r e m a i n i n g 2 0 0
p l a n t s that F u c h s d e p i c t s are n o t f o u n d in the w o r k s o f the
ancients. W h e n he is a b l e to l o c a t e a s p e c i e s that r e s e m b l e s o n e o f
the a n c i e n t t y p e s in o v e r a l l m o r p h o l o g y , he is careful to p r o v i d e an
illustration o f the local r e p r e s e n t a t i v e a n d to n o t e any a p p a r e n t
d i s a g r e e m e n t as to habitus, habit, o r habitat.
T h e naturalistic a p p r o a c h to p l a n t k n o w l e d g e p r o g r e s s e s even
f u r t h e r in Bock. A c o n t r i b u t o r to Brunfels's E i c o n e s a n d a k e e n
c o l l e c t o r a n d o b s e r v e r of plants, B o c k is m o r e attentive t h a n his
p r e d e c e s s o r to d e s c r i b i n g the p a r t i c u l a r i t i e s o f plants "in o u r
G e r m a n y . " 17 F o r instance, he p r o v i d e s the first d e s c r i p t i o n o f the
p a s q u e - f l o w e r , o r H e r b a venti (i.e., A n e m o n e pulsatilla L.), that
" u n k n o w n h e r b " which local folk call " d i n n e r bell: ( K i i c h e s c h e l l ) .
16. Here Fuchs employs binomial nomenclature both for the superordinate
specieme, Cucurbita Sativa, as well as for the subordinate specifics: Cucurbita
Oblonga (Lang Kiirbfs), Cucurbita Maior ( Gros Kiirbfs), Cucurbita Minor (Klein
Kiirbfs). Elsewhere he identifies binomial, trinomial, and even quadrinomial
variants of a type, some of which appear to be speciemes in their own right; as
with Bock, for example, Fuchs has Elleborus Niger Sylvestris sharing a chapter
with Elleborus Niger Adulterinus Hortensis, while Elleborus Albus occupies its
own chapter. This indicates that, at least in some cases, there is a recognition of
affinities that overarch folkgenerics, or speciemes. This is all the more apparent
when affinities are noted between speciemes that are allotted separate chapters
but do not exhibit nomenclatural ties: "The Anethum grows in height up to a
cubit and a half. It has several stems and branches. It has leaves as thin as thread,
a yellow flower. The root, in the form of wood, is not very long. The pompoms
and umbels are like the fennel, with which it shares an almost total resemblance"
(my italics). But such broader notions of affinity do not reflect an overriding
order of relationships or represent absolute standards of awareness in any sense.
17. When deciding between classical authorities or advancing his own
particular views, Bock often appeals to how plants are in nostra Germania.
217
Brunfels had earlier noted the existence of Kiicheschell but chose

not to describe it because it was unknown to the ancients. 18 As
does Fuchs, Bock intermittently refers to more general relations
of affinity between linguistically unrelated kinds, such as the
similarity between the "acrid and fermented taste" of the pasqueflower and that of Ranunculus/9 Furthermore, the sequencing of
chapters manifests an awareness of the intuitive importance of
"family" chains of gramineous, umbelliferous, leguminous, and
labiate herbs. Bock also seems to be well aware that such familylevel chains may cross-cut recognized life-forms: "Rosemary and
lavender are genera of l a b i a t e s . . , but they are s h r u b s . . , therefore the proper place for them is away in Tragus's [i.e., Bock's]
Third Book, among the woody growths . . . but he has both of
these here in the First Book, at the end of the labiates, all the rest
of which are herbs" (Greene 1983, 1:333).
The naturalistic technique of the herbalists reaches its highest
expression with Valerius Cordus. Cordus, who died while on a
botanical excursion to southern Italy in 1544, published nothing
during his short lifetime. His highly accurate descriptions of some
five hundred plants appeared only posthumously in Historia
plantarum (1561). 20 This milestone in descriptive botany was
edited and partially illustrated by the encyclopedic Zurich philologist, physician, and naturalist, Konrad Gesner, who had previously prefaced the Latin edition of Bock. 2~ In each description
one usually finds accurate accounts of the movement, measurement, and arrangement of leaves; the mode of origin and characteristic features of the stem, branches, and flower; the numbers of
loculi in the fruit and rows of the seed; the habit of the root; as
well as the taste and smell of the parts and the sap. The habitat of
the plant also figures prominently, as do notions of qualitative
18. Much of the credit given Brunfels as the father of German herbalism may
actually be due his illustrator, Hans Weiditz. It seems that Weiditz insisted on
figuring even such herbae nudae as the pasque-flower, forlorn by the ancients and
thus only regretfully included by Brunfels (Arber 1953:323; cf. Sprague
1928:113).
l 9. It is notable that Bock, like other German herbalists, relies on nonvisual
as well as visual cues. In this as in other respects, the herbalists' means of
comparison are more akin to those of ordinary folk than to those of the classical
systematists and methodists.
20. The same year saw the appointment of Europe's first professor of field
botany (ostensor simplicium) at the University of Padua.
21. Gesner himself first described the tulip (Tulipa turcaram). Of Persian
origin, the plant was introduced to Europe in the late 1550s by the emperor's
ambassador to Constantinople (cf. Wellisch 1975).
218
SCOTT ATRAN
relationships with other plants; and the mode by which ferns

(trichomanes) propagate is intimated for the first time (cf. Sprague
and Sprague 1939). There is also some (intermittent) suggestion of
a principled "generic" listing based upon a privileged notion of
floral character. Each list is headed by a traditional or familiar
type. For example, a list of twelve Ranunculi includes three
northern European anemones and the pulsatilla; it begins with
Ranunculus palustris (Ranunculus sceleratus), followed by R.
sardous, R. tertia (Anemone ranunculoides), R. quartus (A.
nemorosa), etc. Concern with floral structure also allowed Cordus
to extend the scope of family chains; for example, by thus attaching bryonies to the squashes and melons.
Thus did the Renaissance herbalists conscientiously invert the
focus of scholastic and medieval herborizing, which had tended to
stray from common sense -- that is, from the sort of folkbiological
knowledge that customarily (and cross-culturally) passes for empirical fact. Local knowledge was recovered and precisely transcribed. Only then could it be transcended. A series of interrelated
technological innovations helped to make this possible. In Germany, the invention of movable type and advances in the art of the
woodcut facilitated the dissemination of information about specimens, and these developments were first advantageously exploited
by Brunfels and Weiditz. Thus, regarding Weiditz's illustration of
Weissz Seeblum in Brunfels's herbal, Jacobs aptly notes:
The plant was taken out of the water, and the roots were
cleansed. What therefore we see depicted is a water lily without
water -- isn't this a bit paradoxical? All relations between the
plant and its habitat have been broken and concealed. And yet
this is regarded as the first herbal with illustrations "true to
nature"; Weiditz was a pupil of D/.irer's, and no doubt had
learnt from the master the motto about nature: "Wer sie heraus
kan reissen, der hat sie" -- to tear out nature is to possess her.
(Jacobs 1980:162--163).
In addition, botanical gardens were opened, thus permitting
observation of plants outside of their original environments. The
simultaneous introduction of new techniques promoted-in northern Italy by Luca Ghini for the preservation of dried specimens
in herbaria allowed free exchange and reference back to actual
specimens at all times of the year. Such was the novel heuristic
value of these techniques that Gesner was led to boast that he
could appropriately designate any exotic specimen if presented
219
with the dried flower and leaf. 22 These innovations, together with
the choice of Latin as a c o m m o n tongue, allowed the herbalists to
unambiguously communicate the details of their respective folk
stocks and their exploratory inventories of the flora of neighboring
lands. Henceforth plants could be physically isolated and their
observable structure desiccated for all time. This qualitatively
reduced structure could then be "objectified" in the fixed spatial
perspective of Renaissance art and its "true nature" eternally set in
the neutral tones of scholarly discourse.
III
As a result of the naturalistic m o v e m e n t in herbalism, a worldwide catalog of readily visible forms could be envisaged with
ancient Mediterranean types serving as paragons to which both
temperate and exotic species might be attached. If no ancient type
were available, a m o r e familiar local sort could be substituted. In
this way, speciemes would be m o r e easily c o m p a r e d and contrasted. Local understanding, no matter how provincial initially,
could then be related to a b r o a d e r view of the world. T o this end,
Johann Bauhin 23 and his younger brother Caspar surveyed some
six thousand forms of plants, thus bringing the herbalist period in
western E u r o p e to a culmination. The systematic worth of their
surveys, however, is still a matter of some misvaluation.
Many c o m m e n t a t o r s regard the group headings under each
section of Caspar Bauhin's Pinax theatri botanici (1623), for
example, as reflecting an unmistakable notion of the m o r p h o logical genus, and the numbered forms under each heading as
indicating a solid understanding of the biological species. Some
even regard Bauhin's way of naming plants a s a foreshadowing of
Linnaean binomial nomenclature and his arrangement of groups
into sections, together with the sequencing of sections into books,
as constituting the intuitive foundation for a system of natural
families. But Caspar Bauhin's role as "precursor" in regard to
species, genus, and family is actually much less clear than might
22. Gesner did, in fact, compile a collect,un of some 1500 reasonably

accurate plant illustrations with "dissections" (Gesner 1972), many of which were
based on dried specimens sent from correspondents all over Europe (cf. Gesner
1584). But none were published in his lifetime and only about a third were in
print two centuries after his death (cf. Milt 1936).
23. J. Bauhin trained under Fuchs at Tiibingen, worked with Lobelius under
Rondelet at Montpellier, studied with Aldrovandi, and traveled with Gesner
through the Rhaetian Alps and Val Tellina.
220
SCOTt ATRAN
initially a p p e a r to the historian of biology unfamiliar with folkbiology.

T h r o u g h o u t the Pinax, g r o u p headings within a section are
often m o n o n o m i a l , t h o u g h just as often they are not, and the first
n u m b e r e d f o r m listed in each g r o u p usually c o r r e s p o n d s to the
earliest k n o w n or most easily recognized type of the group.
C o n s i d e r the fourth section of b o o k VIII, which treats only (some
of the then-known) cucurbits, that is, m e m b e r s of the family of
cucumbers, melons, and squashes. T h e g r o u p s (with their types)
are as follows: C u c u m i s (Cucumis sativus vulgaris), Melo (Melo
vulgaris), P e p o (Pepo oblongus), M e l o p e p o (Melopepo clypeiformis), A n g u r i a (Anguria citrullus dicta), Cucurbita (Cucurbita
major sessilis), Colocynthis ( Colocynthis fructu rotundo major),
and C u c u m i s Asininus (Cucumis sylvestris asininus dictus). All
groups have several representatives save the last, C u c u m i s Asininus, which has only its type as representative.
At first glance, there is a striking c o n c o r d a n c e between this
listing and T o u r n e f o r t ' s listing of genera u n d e r the u n n a m e d sixth
section of his class of C a m p a n i f o r m e s in I~l~mens de botanique
(1694). All but one of Bauhin's g r o u p headings lends its n a m e
to a T o u r n e f o r t i a n genus, with Bauhin's first type listed as first
species. 24 Unlike Tournefort, though, Bauhin gives no special
generic descriptions; unlike Linnaeus, he offers no familial designations; unlike Ray, he mentions no characters that might distinguish species as such. In fact, the resemblance between the
fourth section of b o o k V I I I and the first eight sections in b o o k X X
of Pliny's Historiarum mundi is just as marked. Yet, Pliny could
scarcely be taken for a p r e c u r s o r of Ray, Tournefort, or Linnaeus.
Indeed, all but two of Bauhin's g r o u p headings c o r r e s p o n d to
section headings in Pliny. 25 F u r t h e r m o r e , while Pliny generally
treats but one generic-specieme in a section, it is that which
Bauhin takes as his first type. 26
A l t h o u g h Pliny seldom discusses m o r e than one or two forms
24. The single difference concerns Bauhin's chapter, Cucumis Asininus,
whose sole species in included under Tournefort's genus Cucumis.But Linnaeus
would again accord it a status distinct from the cucumbers, i.e., as Momordica
elateriumL. (elateriumwas the name given it by Pliny).
25. Neither Melo nor Melopeppo appear as section headings in Pliny.
Nevertheless, in bk. XIX, sec. 23, which deals with some of the general properties
of cucumbers and melons, Pliny remarks that a "new form" (nova forma) has
been found in Campania whose fruits are called melopeponas.
26. The one exception to the rule is in bk. XX, section Cucurbita. Pliny
discusses only a wild variety, while Bauhin's type refers to one of the two
domestic varieties Pliny discusses elsewhere 09k. XIX, sec. 24).
221
per section, whereas Bauhin sometimes names as many as twenty

distinct forms for each group, no principled differences in the use
of nomenclature are apparent. Moreover, Pliny explicitly notes the
(family) similarity that squashes have with cucumbers and melons
(similes et cucurbitis natura). In fact, there are no principled
differences between folk nomenclature or understanding of family
relationships and Pliny's or Bauhin's. Tzeltal knowledge of cucur o
bits is a case in point. The Tzeltal recognize, but do not label, a
"covert" category of cucurbits that encompasses eight genericspeciemes (Berlin, Breedlove, and Raven 1974:421--424); six of
these are native squashes and gourds (c'um, mayil, c'ol, bohc, cu,
c'ahko?), and two are Old World imports (melones, santiya). 27
Except for the gourds bohc and cu, which are cultivated subspecies of Lagenaria siceraria, each generic-specieme corresponds
to a distinct scientific species. The majority of generic-speciemes
each subsume two or more binomially labeled folkspecifics, with
"pumpkin" (c'um), for example, having perhaps as many as ten:
"soft pumpkin:" (sk'un c'um), "hooked-neck pumpkin" (luk' nuk'
c'um), "tree pumpkin" (ste ? c'um), and so forth. Longer polynomial forms are absent from this covert Tzeltal category, but
trinomial and even quadrinomial forms, though much less frequent
than binomials, do appear in other categories.
The tendency, for the sake of convenience, to regard only such
Bauhinian groups as appear to presage the genera and families of
future taxonomists has thus led commentators astray; for these
ostensibly "standard" groups exhibit no principled distinctions
from ancient or folk groupings. Examining some of Bauhin's more
"problematic" groups, though, one finds far less regularity in
nomenclature or appreciation of morphological affinity. These
groups, which actually compose the majority of cases, differ
appreciably from ancient (and folk) as well as future taxa. Often,
the groups of a section exhibit no regular pattern of nomenclatural
association or dissocation with one another; and the listed species
of a group may bear no linguistic tie at all with the group heading.
For example, the second section of book VIII lists the following
groups: Clematitis, Clematitis Indica, Clematis Daphnoides, Pericymenum, Apocynum, Asclepias, Polygonatum, Lilium Convalium, Monophylum, and Laurus Alexandrina. Under the group
Clematitis Indica, seven forms appear: the first three forms have
27. Tribe, Cucurbitae. Subtribe: A. Benincasinae -- Lagenaria siceraria
(oohc, cu), Citrullus vulgaris (santiya); B. Cucurbitae -- Cucurbita ficifolia
( mayil ), C. pepo ( c 'ol ), C. moschata ( c'um ).
Tribe, Melothrieae, Subtribe, Cucurminae: Cucumis melo (melones).
222
SCOTt ATRAN
polynomial structures that begin with the binomial indicating the

group heading, namely, Clematis indica . . . ; the fourth and fifth
forms begin with the binomial Clematis malabarensis; and the
last two forms have names that in no way link them with the other
forms of this or related groups of Clematitis: Colubrini lignitertium
genus in eadem provincia and Radex quaedem in Malaca.
T o make matters worse, many groups are notoriously heterogeneous. Take the very first section of b o o k I concerning grasses"
(De graminibus). Nearly all of the twenty-four groups in the
section have binomial, rather than mononomial, headings. Three
groups include subgroups whose headings are usually trinomial in
structure. T h e large majority groups as well as subgroups of the
section cross-cut two or m o r e Linnaean genera. F o r example,
the subgroup " G r a m e n paniculatum arvense" has four n u m b e r e d
forms, each of which corresponds to a Linnaean species of a
different genus: Gramen arvense panicula crispa ( Poa bulbosa L.),
Gramen panicula pendula aurea ( Cynosurus aureus L.), Gramen
segetum altissimum panicula sparse (Argostis spica-venti L.), and
Gramen segetum panicula arundinacea (Aira cespitosa L.). 28
Yet it would be a mistake to see this lack of regular nomenclatural or morphological pattern as a return to the debased
practices of late antiquity and medieval scholasticism, or as a
deviation f r o m the orderliness of folk classification and earlier
Renaissance herbalism. In those cases where many m o r e forms are
known to Bauhin than to ordinary folk or to ancient and Renaissance herbalists, his comprehensive attempt to index all known
plants leads to heterogeneity in nomenclatural and morphological
affiliation. This is not because he abandons c o m m o n sense, but
because c o m m o n sense alone is not intrinsically ablb to relationally segregate basic forms from one another when most of
those forms originate in entirely different environments, and when
the n u m b e r of forms exceeds that of a normal locale by an order
of magnitude.
In fact, the various nomenclatural combinations of the Pinax
express rather consistent extensions of c o m m o n - s e n s e naming and
segregating practices to problems that ordinary folk are not
normally compelled to deal with. Only occasionally are folk
required to attach an unfamiliar type to the basic stock of familiar
28. A notable exception to such morphological heterogeneity in the section is
the group Gramen Paniceum, which has five numbered forms corresponding to
three species of a single Linnaean genus (Panicum); another is the subgroup
Gramen nemorosum hirsutum, which has six numbered forms corresponding to
four Lirmaean species of the genus Juncus.
223
types. For Bauhin, however, both locally familiar and ancient types
were few and far between, and the whole of the then-known living
world could not be commonsensically accommodated to these
except in a fragmentary and piecemeal fashion. If, to a modern
eye, family-level and generic-level chains of species can be discerned in Bauhin -- the most accomplished of the herbalists -- for
him these is as yet no clear distinction between, nor conception of,
species, genus, or family.
Bauhin's Pinax, or "Chart," was little more than a catalog of
forms and synonyms for the plants then known, which included
not only plants of southern and northern Europe, but also many
from the better explored coastal areas of Africa, Asia, and North
and South America. The Pinax did not provide a key to understanding the living world at large, although its practical value for
cross-referencing was unrivaled for the better part of two hundred
years. There was nothing in it to allow a systematic placement of
kinds not originally surveyed in a fixed relation to known kinds.
The systematists would seek to remedy this by logically codifying
readily visible patterns of morphological affinity and dissimilarity.
Because exploration was constantly yielding new kinds, a sure
code would have to be automatically extendable so as to permit
one to tabulate the place of any given local form of plant with
respect to every other (possible) form in advance of all future
discoveries.
IV
It was Andrea Cesalpino, professor at Pisa and Papal physician,
who first suggested a way whereby the intellect might detach itself
from the overwhelming confusion of so many new sensible
forms. 29 Yet, even Cesalpino's most novel contributions to the
development of systematics have been severely misconstrued for
lack of appreciation of the singular character of his "Aristotelianism." T o be sure, Cesalpino's Questionibus Peripateticis (1571)
constitutes a landmark in Aristotelian exegesis of the late Renaissance. But the botanical applications of the conclusions of this
exegesis in De plantis libri X V I (1583) are designed to resolve em-
29. There are suggestions in Gesner's letters that he was thinking of some of
the crucial issues concerning the delimitation of species and genera in ways
similar to Cesalpino; however, these ideas were never systematically developed
and their direct influence on the subsequent course of natural history appears to
have been marginal.
224
SCOTT ATRAN
pirical problems quite different from those envisaged by Aristotle,

and wholly alien to the nonempirical concerns of the scholastics.
There are two principal and interrelated features of "Aristotelian essentialism," as the creed has come to be known among
historians and philosophers: the dogma of the eternal fixity of
species (Hull 1965), and the doctrine that any individual necessarily comes to be what it is in virtue of its species-specific
properties, that is, those properties which define the essence of the
kind of being that that individual is and which therefore make the
individual the particular individual it is (Quine 1966:173--174). It
is important to note that while these features do characterize
Cesalpino's essentialism, they do not appropriately apply to
Aristotle. Moreover, the method of Logical Division employed by
Cesalpino differs significantly from Aristotle's own method. This is
not because Cesalpino "equivocated" in his use of the Stagirite's
principles in order to reconcile a priori ideas with reality (Sachs
1890 [1875]), or because he was forced to make "concessions"
in the analysis of material conditions (Bremekamp 1953), but
because he was faced with an organizing task markedly distinct
from anything encountered by Aristotle.
Aristotle's essentialism effectively boils down to an acceptance
of common-sense animal universals -- that is, the generic-specieme
(atomon eidos) and life-form (megiston genos) -- as "natural" and
"for the better." Given the pretheoretical array of basically distinct
kinds in the local environment, Aristotle seeks to understand what
is good in their continuance over time, their diversity, and their
ultimate connectedness in nature's overall plan. Because all life is
perishable, continuance can only be renewal -- a constant cominginto-being. The renewal of any particular kind is conditional upon
three factors: the availability of the right sort of matter, the
reliability of material forces, and the presence of form-potential
(dynamis) for combining matter in kind with the aid of material
forces. The female menses (catamenia) are responsible for the
first, the sun, wind, climate, and available nutrients for the second,
and the male seed for the third. The third factor is determinant
(aitia kai archai) inasmuch as it conveys life-giving soul (psyche).
This form-potential is invariably in the father's image; however,
the actual form that is realized is always a "falling-short" (steresis)
of the father's likeness. At best, then the realized offspring can
only be said to bear a "species-likeness": "since mortal things
cannot share continuously in the eternal and divine (because
nothing that perishes can preserve its identity nor remain numerically one), they partake of eternity and divinity in the one way that
is open to them, and with unequal success; achieving immortality
Origin of the Species and G e n u s C o n c e p t s
225
not in themselves, but vicariously t h r o u g h their offspring, which,

though distinct individuals, are o n e with t h e m specifically (eidei)"
(De anima 4 1 5 a 2 6 - - b l ) . Thus, individuals strive to perpetuate
themselves t h r o u g h sequences of ancestor-generating-descendent.
But there is no indication whatever that such a self-generating
lineage is ever eternal or immutable in fact. T h e only "species"
(eide) that Aristotle mentions as even a p p r o a c h i n g eternity are
such subcategorical "genera" (gene) of substance (ousia) as man,
animal, and plant: "that is why there is always a kind (genos) --30
of m e n and of animals and of plants" (De generatione animalium
73 l b 2 5 - - 7 3 2 a 2 ) . 31
It is not the case that species exist "fixed and unchanging"
(Lloyd 1968:88) ''from all eternity . . . by the process o f like
generating like" (Sloan 1972:2). Rather, natural species represent
temporally and spatially localized populations, the individuals of
which bear a resemblance in the process of genesis - - a process
characterized by a range of active (male) form-potentials, passive (female) material-potentials, and intrusive (environmental)
element-potentials (Atran 1985b). T h e species is a naturally
occurring empirical "necessity" - - part of nature's ontological fold
that is nevertheless conditional u p o n an ideal constellation of
material circumstances that m a y never, in fact, obtain (cf. Gotthelf
1976; L e n o x 1980; Balme 1980). Such necessity is not eternal,
because n o r m a l conditions for generation and growth m a y shift by
c h a n c e (automaton), and nature's optimal course along with them.
-
30. Eidos and genos appear as interchangeable terms in Aristotle's oeuvre,

except when reference is made in the biological works to the folkgenericspecieme (atomon eidos) and life-form (megistongenos).
31. Man, animal, and plant may be counted among the principal ontological
subdivisions of substance. As such, they are intimately bound to the structure of
language (as basic semantic categories) and the organization of the perceptual
and conceptual framework through which we apprehend the world (as fundamental regions of object-giving intuitions). Indeed, given Aristotle's commitment
to common-sense realism, it is not surprising to find cross-cultural evidence that
people take them as basic ontological types (cf. Kesby 1979) whether or not
single words are given them (cf. Brown 1984). Even children appear to use them
from the outset for structuring their language: so that some predicates, such as
"die" and "get sick," sensibly -- that is, truly or falsely and nonmetaphorically -apply to all and only living things; predicates like "feel" and "have babies" apply
exclusively to humans and animals; while "blossom" and "wilt" apply only to
plants, and "worry" and "joke" only to humans (Keil 1979; cf. Aristotle,
Categoriae lb16). Such types also appear to provide frameworks for induction.
Thus, if two appreciably different animals are found to have a property in
common (e.g., if it is known that worms and horses eat), then adults and children
alike will initially tend to attribute that property to all and only animals (cf. Carey
1978).
226
SCOTT ATRAN
N e c e s s i t y is c o n d i t i o n a l , c o n s e q u e n t i a l , a n d after t h e fact. It is n o t
absolute, a priori or causative in the sense of being directed by a
p u r p o s i v e f o r c e like G o d . 32
So, f o r A r i s t o t l e , o n l y w h e n c e r t a i n e n v i r o n m e n t a l c o n d i t i o n s
o b t a i n c a n t h e t y p i c a 1 morphe e m e r g e i n a g i v e n i n d i v i d u a l w h o s e
u n d e r l y i n g n a t u r e falls w i t h i n a g i v e n r a n g e . 33 T h e m o r p h o t y p e ,
t h a t is, t h e v i s i b l e f o r m o f t h e species, is n o t itself a n a t u r e (physis),
n o r is it a l w a y s r e a l i z e d f o r i n d i v i d u a l s w h o s e n a t u r e falls w i t h i n
a g i v e n specific r a n g e . T h e m o r p h o t y p e e m e r g e s o n l y "for the
m o s t p a r t " a n d " b y n a t u r e " (physei) as a n o p t i m a l c a r e e r u n d e r
certain material circumstances. Although Aristotle applies the
t e r m " n a t u r e " to (i) t h e u n d e r l y i n g p o t e n t i a l for d e v e l o p m e n t in
t h e i n d i v i d u a l , (ii) t h e a c t u a l p r o c e s s o f d e v e l o p m e n t , a n d (iii) t h e
final r e s u l t o b t a i n e d , e a c h o c c u r r e n c e o f t h e t e r m d e n o t e s a diff e r e n t o n t o l o g i c a l e v e n t (Physica II, i). T h e s e d i f f e r e n t e v e n t s a r e
closely r e l a t e d , b u t the r e l a t i o n s h i p is i n p a r t g o v e r n e d b y m a t e r i a l
c o n t i n g e n c y . T h i s is n o t t h e c a s e for C e s a l p i n o , w h o c o l l a p s e s all
t h r e e e v e n t s i n t o a single e x p r e s s i o n o f D i v i n e c o n t e m p l a t i o n .
32. Although the characteristic attributes of both eternal and sublunary kinds
are necessary, they are necessary in different ways. In both cases, the full
complement of such attributes includes the essential traits as well as those
derivative traits that naturally accompany the essential ones as "proper" consequences following from the intrinsic nature of a thing. (A proper trait may be
unique and necessary to the species, but an essential trait must also pertain to the
genus; that is, each of the other species of the genus must have as an essential
difference a complementary, or coordinate, feature taken from a feature-dimension that spans the whole genus.) But for mathematical eternals, necessity is
absolute and cannot be otherwise (Metaphysica 101567, 102667), while for
natural kinds necessity is merely hypothetical and follows only "for the most
part:" (1027a f.). In other words, necessity is conditional upon the actual comingabout of the end state. Thus, on the hypothesis that the mature organism (acme)
will fully develop, then what is anatomically and morphologically typical of,
and proper to, that species of organism will fully come about only because the
"best possible" end state has come about. There is no guarantee that the best
possible end state will be achieved in fact, but under normal circumstances it
generally does tend to come about (De generatione animalium 731620 f., Physica
199b15 f.).
33. Thus, given a range of compatible male-females paris of underlying
individual natures and an environment conducive to the proper functioning of
individuals with those natures, then it is only usually the case that individuals will
resemble one another as to type. Or, to put it another way: given that individuals
resemble one another as to type, then it is necessary that their natures'fall within
a specific range and that certain environmental conditions obtain. This allows
that radical environmental change might lead to changes in the structure of
coexisting communities. Conversely, should conditions severely restrict the
movement of individuals whose natures do not normally fall within bounds of
compatability but are not entirely incompatible, then new communities may
appear if such conditions remain constant (e.g., around desert oases).
227
For Aristotle, as for Cesalpino, the form (eidos), or soul

(psyche), of the individual organism is responsible for its physical
organization and vitality (De anima 415b9--30). But in Aristotle's
case, the life-giving principle (archai) and immanent form (dynamis) are shared only by a sire and its immediate progeny.
Material happenstance (from insufficiency in the coction of vegetative female matter to the heat of the south wind) unavoidably
deflects the unfolding of the causal process from its ideal course,
which is the actualization of the sire's form. By default, the
progeny tends to resemble the more weakly determined form of
one of its ancestors. If no individual form predominates, "there
remains what is common," namely, the morpho-geographic species
(generic-specieme) as a limit at which a given range of female
matter and male form are compatible (De generatione animalium
768b10 f.). Because the form-as-soul that the progeny in turn
passes on to its offspring is determined by the former's actual form
-- not the original sire's -- there is little, if any, chance for a
perpetuation of any given form across generations. The immanent
forms that are passed along to individuals of a species are similar,
but not identical.
For Cesalpino, though, species are not unrealized optimal
tendencies for individual development in the material conditions of
nature sub specie universalitatis. They are everlasting, immanent
forms for the material realization of God on earth sub specie
aeternitatis: "Eternity can only arise from the eternal: since the
proper work of the vegetative soul is to engender its like, which
makes for the eternity of the species, it is necessary that its
substance not be corruptible. The reason for the eternal lies
neither in corruptible existences taken individually, nor in their
totality" (1571:11, viii). The environment really has no formative
role to play. It can but facilitate or hinder the only natural,
preordained sort of development possible, namely, that of an
underlying nature of specific kind into the mature form of a
specific kind. The underlying natures that are housed in the seeds
of the various individuals of a species are not merely members of a
range of underlying natures that are similar in potential; they are
identical "ex semine ad specierum aeternitatem" ( 1583:1).
Cesalpino adopts much of Aristotle's realist epistemology (cf.
Dorolle 1929:22--23) and embraces Aristotle's instantiation requirement: contrary to the Platonist position, no form can exist
without being materially individuated. But Cesalpino adds a heavy
dose of Christian theology, which results in an unavoidably neoPlatonic slant. For both Cesalpino and Aristotle, human intelligence apprehends its objects rather than constructs them. More-
228
sco'rr ATRAN
over, the world that appears to us -- that is, that which is visibly
manifest -- is fundamentally the only world that is. Intelligence
extracts the universal aspects of reality apparent in individual
bodies by a sort of "induction" (epagoge). Just as the vegetative
soul of the organism is embodies in its whole morphology, so the
intellectual soul of the human being is enmattered in its various
sensory images of the world. But these images are themselves the
disembodied presentations of the individual forms of particular
organisms to the human mind. The mind, which is endowed with
light-sensing capacity (when considered potentially), and lightgiving quality (when actually operating), illuminates the clear,
distinct, and common idea in its several images. In the mind, then,
"actual knowledge is identical with its object" (De anima 430a10-25), and through the mind objects become, as it were, conscious of
themselves (Cesalpino 157 l:II, viii).
In this way, the mind grasps what is universal in the individual
(universalia in rebus). Whatever is thus universally intelligible,
constitutes an immortal part of the cosmos. But for Aristotle,
everything that occurs in the sublunary realm occurs as a persistent tendency that always falls short of perfection; whereas
for Cesalpino, this unremitting emulation is itself immanent
perfection. In the one case, the continuous historical cycle of
ancestor-generating-descendant is but analogous to the fixed
rounds of the heavens; while in the other, abiding sublunary cycles
are, ultimately, identical with the everlasting celestial circuit.
In Cesalpino's Christian cosmos, all movement must be considered an immanent striving, and ultimate attainment, of eternity
and oneness with a perfectly whole, contemplative, unmoved God.
Accordingly, Cesalpino holds with Albertus that the form-potential of the species is perfectly perpetuated over its individuals
("generatio naturalis terminatur ad perfectum" [Albertus Magnus,
Quaestiones super de animalibus XVI, 20]). It survives death and
decay that it may achieve eventual redemption. With Albertus's
student Thomas, Cesalpino also maintains that the individual
form-potentials of a natural lineage of organisms are exactly the
same "in species," although actual forms do manifest "accidental"
differences owing to material deficiency or to the progenitor's
weak generative virtue ("generans generat sibi simile in specie:
fit tamen aliquando aliqua dissimilitudo generantis ad genitum
quantum ad accidentia, vel propter materiam, vel propter debilitatem virtutis generativae" [Aquinas, Summa Theologiae III, 74,
iii]). A species is thus a nature in continuous action.
As with the object of knowledge, so with knowledge itself:
knowledge of universals is perfect, true, eternal, and universal
O r i g i n o f the S p e c i e s a n d G e n u s C o n c e p t s
229
k n o w l e d g e . O n A r i s t o t l e ' s a c c o u n t , though, to the e x t e n t that a

s p e c i e s (i.e., a n a t u r a l lineage) exists in t h e w o r l d , it d o e s so only as
an a g g r e g a t i o n o f i n d i v i d u a l s similar, b u t n o t identical, in b o t h
actual f o r m a n d f o r m - p o t e n t i a l . M o r e o v e r , i n s o f a r as t h e s p e c i e s
exists in the m i n d , it offers n o g u a r a n t e e o f c e r t i t u d e , b u t o n l y o f
the l i k e l i h o o d a n d verisimilitude. 34
In the A r i s t o t e l i a n c o s m o s , p h y s i c a l f o r c e s o p e r a t e o n l y a m o n g
the f o u r e l e m e n t s in the s u b l u n a r y r e a l m , while t h e c h a r a c t e r i s t i c
m o t i o n o f celestial b o d i e s is g o v e r n e d exclusively b y t h e d e m a n d s
o f g e o m e t r i c a l p e r f e c t i o n . C e s a l p i n o b r e a k s d o w n this d u a l i s m not,
as G a l i l e o w o u l d , b y p o s t u l a t i n g that p h y s i c a l causality p e r m e a t e s
the e n t i r e c o s m o s , b u t b y m a i n t a i n i n g that the s u b l u n a r y r e a l m
also attains a m a n n e r o f p e r f e c t i o n . This p o s i t i o n allows C e s a l p i n o
to e l a b o r a t e a n o v e l c o n c e p t i o n o f n a t u r a l h i s t o r y a n d t a x o n o m y .
H e goes e v e n f u r t h e r t h a n the h e r b a l i s t s in w r e n c h i n g the o b j e c t o f
i n q u i r y f r o m its actually p e r c e i v e d state o f n a t u r e . S e n s o r y assessment, o t h e r t h a n the a u s t e r e b l a c k - a n d - w h i t e r e p r e s e n t a t i o n o f
vision, is virtually b a n i s h e d f r o m all d e s c r i p t i o n , as a r e c o n s i d e r a t i o n s o f e c o l o g i c a l situation. F r o m n o w on, the v a r i o u s
i n t e r a c t i o n s b e t w e e n h u m a n s , animals, a n d p l a n t s c o u l d b e effectively i g n o r e d at all g e n e r a l levels o f s t a t e m e n t .
Thus, all o r g a n i s m s h a v e u n d e r l y i n g n a t u r e s that c a u s e g r o w t h
a n d d e v e l o p m e n t , just as t h e y have m a t u r e a n d d e v e l o p e d s h a p e s
34. A difficulty that agitated the schoolmen concerns the dual nature of
forms and universals. Forms exist in nature (in rerum natura) only as a many of
actual individuals and as a potential universal. But in the mind, form appears to
exist as a one, that is, as an actual and unitary universal. If, however, universalsin-the-mind are individual thoughts they cannot be truly universal, for they are
embodied in individual thinkers. Communication of ideas and consensus become
contradictions in terms, truth and knowledge only illusions. The resolution of this
paradox seems not to have especially preoccupied Aristotle (cf. Skulsky 1968).
Cesalpino opts for the Thomist solution (cf. Aquinas's Tractatus de unitate
intellectus contra averroistas). Men's minds are the same "in species" but not in
number. The mind is immaterial. Yet, it can only come to be through embodiment, just as an individual idea is separate form those sensations that initially
occasion and confirm awareness of the idea. Once the body dies, though, the
mind can survive, just as ideas can operate independently of their original
"triggering" experiences. The mind, with its ideas, produces a "likeness"
(similitudo) of external reality. This picture represents reality under its eternal
aspect, and not as seen through fleeting sensory images (phantasmatibus) of
transient experiences. People do not actually share representation, but they have
representations of the same things. What people do share is the potential to grasp
the world as it truly and always is. This potential is an a priori power of the
"active intellect" (intellectus agens) to sort out sensory experience according to
the natural order, that is, by defining natural kinds. It is through this God-loving
power to contemplate the eternal order that the mind becomes "part of" that
order (Cesalpino 1571 :II, viii).
230
SCOTT ATRAN
that are caused by those natures. These mature shapes are direct
manifestations of G o d on earth, which it is the privilege and duty
of m a n to contemplate. T h e r e are different kinds of natures, systematically related to each other, as there are distinct kinds of
shapes between which the geometrician can perceive the formal
relations. Because natures are the causes of shapes, the formal
relations that hold between natures - - o r their perceptible representatives and repositories (i.e., seed structures) - - should themselves represent (as simulacra) the eternal relations that hold
between shapes (i.e., perceived species) in the mind of the Divine
Geometrician. T h e result would be a formal analysis of internal
causal relations (substantia) (Cesalpino 1583:26). This would
enable one to understand "the n u m b e r and nature of principles"
(1571:1, i) underlying the p h e n o m e n a revealed by sense perception (cognita). By resorting to a formal analysis of causes (i.e.,
of natures), o n e could consistently maintain an attachment to
the Aristotelian episteme without subscribing to the Aristotelian
m e t h o d of functional analysis, that is, without having to treat the
d e v e l o p m e n t of natures u n d e r specified material conditions.
A s for the task of constructing a universal taxonomy, this, it
would appear, could best p r o c e e d by ignoring local circumstance.
Only in that way might one h o p e to reduce the chaotic multiplicity
of sensible forms to equivalence classes whose p r o p o r t i o n s the
mind could o n c e again manage. F o r Aristotle, classification aims
to provide a c o m p l e t e division (diaeresis) and assembly (synagoge)
of all antecedently k n o w n basic-level kinds. T h e differentiae of
such a classificatory structure must yield a progressive reduction
of habitual, basic-level structures to their essential, functional
parts. But before such a reduction could be systematically engaged, a complete knowledge o f the habitual structures ~of all basic
kinds would be needed. Given the fact that Aristotle acknowledges
a b o u t the same n u m b e r of basic kinds as any folk naturalist, it
would seem to him plausible that such a precondition for classification could be met. 35 But Cesalpino faces a very different task.
35. Aristotle's task was to find a principle of unity underlying the diversity of
ordinary phenomenal types. That meant a systematic derivation of the genericspecieme, or atomon eidos, from the folk life-form, or megiston genos. This first
sustained scientific research program differed from modern science in its preoccupation with explaining the familiar and known (Theophrastus, Historia
plantarum I, ii, 3--4), rather than with exploring the unknown for its own sake.
This program failed owing to a fundamental antagonism between what were
effectively nonphenomenal means and the phenomenal end sought. To explain
the visible order of things Aristotle had recourse to internal functions; but such
functions cannot be properly understood if, as with Aristotle, they are referred
231
For Aristotle, it is first necessary "to know all the species that
fall under the genus" (Analytica priora 68b27), "like sparrow or
crane and all" (De partibus animalium 644a25--30). Whereas for
Cesalpino, one needs to know the genera before all of the indefinitely many species; for, "if there is unclarity in genera, species
will necessarily be confused in many ways" (1583:25): "One must
unite everything within homogeneous genera . . . [which is] very
efficacious for the memory, because an enormous number of
plants are thus enclosed in a rrsumr, ordered by genera; hence,
although a plant may have never before been seen, anyone can
place it in its appropriate category; and if a plant has no name,
anyone can call it by its generic name" (1583, dedicatory epistle).
But before the right genera could be sought for reducing the
multiplicity of new forms to tractable equivalence classes, it would
be necessary to fix a criterion for the species even in advance of
future discoveries. Without such a criterion there could be no
principled justification for uniting basic-level sorts originating in
different climes within the same genus. Such a criterion must,
therefore, establish that morphological characters usually perceived to be constant are, in fact, those that ought to be constant
according to God's eternal plan. This allows one to accept that
(variations in) material conditions are irrelevant to the existence of
species-types. Consequently, one is justified in extracting folk
species from the environments in which they were first located,
and converting them into abstract types that may be placed in a
universal taxonomy.
Accordingly, the species criterion, though often attributed to
Ray (Stearn 1957:156; Mayr 1982:256), or known as the "Linnaean species," is actually first introduced in De plantis:
Plants that resemble one another in the totality of their parts
generally belong to the same species. In effect, a small difference between plants is not always a sign of species difference; often the leaves, flowers, and other parts are modified
by diversity of location and conditions of c u l t i v a t i o n . . , if one
sows the seed of a domestic species, wild plants will readily
arise that are as indifferent in species as those which by cultivation or other means have been modified: for, like everywhere
primarily to their morphological manifestations. Moreover, because Aristotelian

division is geared to an analysis of antecedently known kinds, Aristotle clearly
could not foresee that the introduction of exotic forms would undermine his
quest for a logical order after the fact.
232
SCOTT ATRAN
engenders like, according to nature and of the same species.

(Cesalpino 1583:26).
The local problem of accounting for token variation from the
morphological type, which had so agitated Aristotle, could thus be
reduced to the test of seeing whether a plant comes true to seed.
Following Harvey's (1981 [1651]) axiom that plant and animal
seed are conceptually and ontologically of a sort (omnia ex ovo),
Ray (1686:40) would extend Cesalpino's insights to the zoological
realm (Sicut enirn in Animalibus). For the whole of the living
world, then, "every kind hath its seed" (Ray 1691:181 ).
V
Much opinion has it that Cesalpino's bold initiatives to solve what
must have seemed to the herbalists an intractable problem were
doomed by his Procrustean effort to fit a new world into an
Aristotelian framework inadequate even for the old. It is for his
allegedly outworn
belief in the value of seeds and fruits in supplying criteria for
classification that Cesalpino is most frequently remembered.
This principle, and that of the prime importance of the division
into trees and herbs, to which he also adhered, were direct
deductions from the views of the Aristotelian school . . . . The
empirical development of taxonomics at the hands of the
Renaissance herbalists gave surer guidance than Cesalpino's a
priori views. (Arber 1950:31--32; cf. Cain 1959:234)
This view of Cesalpino's contribution to systematics is gravely
flawed.
First, the Renaissance herbalists gave little, if any, guidance
to the development of taxonomies. Their listings were either
alphabetical or unconnected sequences of implicit folk groupings.
Cesalpino not only captures most, if not all, of the intuitive
regularities of folk and Renaissance herbalists, but seeks to analyze
them morphologically and logically. Renaissance herbalists scarcely
attempted anything of the sort.
Second, the partition of flora into tree and herb, though it may
be justified post hoc by Aristotelian or (as in the case of Ray and
Tournefort) non-Aristotelian reasoning, most certainly did not
originate as "a direct deduction" from any set of a priori principles.
This fundamental partition into what ethnobiologists call "lifeforms" is common to tribal societies the world over. It charac-
233
terizes ancient Hebrews, Chinese, and Mesoamericans, as it does

the lay classification of modern Israelis, Americans, and Tanzanian
tribesmen who could hardly be taken for devout Aristotelians
(Atran 1985a).
Third, Cesalpino's appreciation of the value of fruits and seeds
differs radically from Aristotle's. For Aristotle (and Theophrastus),
as for Cesalpino, the seed and fruit are potential bodies of a
certain type: vital loci wherein actual material forms are generated
by heat (coction) from their underlying natures. In Aristotle,
however, the species-type of the mature, realized organism is not
determined exclusively by the seed, but only in conjunction with
environmental conditions normally compatible with the functional
adaptation of organisms whose underlying natures fall within a
specific range. Although Aristotle, like Cesalpino, believes most
plants to reproduce asexually, he still allows the environment
to have a formative influence on the optimal morphology of the
individual and hence on the composition of the species; Cesalpino
does not. Nor does Aristotle (or Theophrastus) express the teleological thesis of the vital unity of the organism in terms of its
reduction to an essential part. Other parts of the plant may come
into play "for the sake of" the fructification -- which is the plant's
end -- but a true classification must mention and make explicit this
teleological character of the other parts.
Cesalpino's rationale for his reductive thesis derives from the
dual role that he accords the characters of the fructification in the
natural scheme of things. The natural scheme represents a contact
point between God's eternal, rational order and the actual, imperfect, material conditions obtaining on earth. The fructification
constitutes a natural intersection of these orders. It provides God
substantive access to the material world -- for the plant just is the
material actualization of the fructification; and it permits humankind epistemic access from the material world to God. The
taxonomic order is thus identical neither with the eternal order
(conceptually whole and materially undifferentiated) nor with the
sensible order (how things commonsensically appear to us under
their various material guises). It is for this reason that the fructification elements chosen as the differentiae of classification need
have no direct functional value, only rational value. Direct functional value pertains to the sensible realm exclusively. This is not
Cesalpino's fundamental preoccupation; rather, he is concerned
with the rationality imposed by God on the material being. That is
why only the number (numerus), position (situs), and figure
(figura) of the seeds and other selected floral organs are truly
essential to taxonomy.
234
SCOTT ATRAN
H a v i n g s e t t h e c r i t e r i o n f o r f i x i n g s p e c i e s , 36 C e s a l p i n o c o u l d g o
on to establish genera that would bring the chaos of an everexpanding world of sensible forms back to some kind of intellectual order. He argues that the characters (differentiae) required
for classification must pertain to the (morphological) nature of the
p l a n t o n l y . V a r i a b l e c h a r a c t e r s t h a t a l t e r w i t h c l i m a t e , soil, o r
h a b i t a t , o r t h r o u g h m a n ' s i n t e r v e n t i o n , m u s t b e t a k e n as " a c c i dental" and unfit for demarcating the essential similarities and
differences between plants. Qualities of texture, odor, taste, and
36. Gesner may have preceded Cesalpino in recommending the experimental

test of seeing whether a plant came true to seed in order to determine which
characteristics are really specific and which are accidents. It is doubtful, though,
that Ray was aware of this when he proposed the test as an empirical criterion
for the species (1686:40--43). It is possible that Gesner influenced Cesalpino
through correspondence with Ghini, Cesalpino's teacher; or it may be that Ghini
himself conceived the test. It is also worth noting that just as Gesner appreciated
Ghini's technique of drying plants, so, too Cesalpino (1583, dedicatory epistle)
expressed a debt to Ghini's concept of the botanical garden.
According to one influential current of thought in the tradition of the French
epistemologist Gaston Bachelard, knowledge of the living world barely changed
from ancient times through the sixteenth century. On the one hand, there was the
obvious visual diversity and integrity of living kinds. On the other hand, though, it
was thought along hermeneutical lines (exemplified in the Paracelsian school)
that superficial resemblances belied a secret network of invisible forces that
linked each kind to every other as well as to all other objects in the universe.
Biological forms from Aristotle (Joly 1968) to Gesner and Cesalpino (Foucault
1970 [1966]) were thus supposedly imbued with magical causes and mythical
images. As Jacob tells it: "amid this tangle of forms, there is no place for the
species, as this is to be understood in the seventeenth and eighteenth centuries -namely a persistence of visible structure through succeeding generations" (1973
[19701:23 ) .
The source of this tale is the false assumption that prescientific thought is and
was culturally idiosyncratic and capricious, while the proto-scientific thought of
the seventeenth century broke with the "specious" knowledge of similitudes
"against natural training, against the colored and diverse fact . . . by purifying
natural substances and arranging tangled phenomena" (Bachelard 1980:23). On
this account -- as indeed in the empiricist accounts that the Bachelardians are at
pains to refute -- science emerges by purging "stone-age metaphysics" of vague,
sentient images and overhasty, inconsistent associations. The difference between
Bachelardians and empiricists (Sachs 1890; Gilmour and Waiters 1964) on this
score is that the former see an "epistemological rupture" between science and
common sense, while the latter concede a certain measure of epistemic continuity
between the savage and the savant. But, in point of antropological fac~, ordinary
common-sense kinds are not initially symbolically construed, and the taxonomic
relations between them are invariably consistent. Rationality is not a historical or
cultural construct, but a cognitive universal.
Symbolic speculations, though, are culturally idiosyncratic attempts to go
beyond the immediate and manifest limits of common sense. So, symbolic
thought is postrational, or at least posterior to common sense, which is rational.
Origin of the Species a n d G e n u s C o n c e p t s
235
c o l o r are particularly subject to variation a n d c a n b e used only as

adjuncts to the a u s t e r e visual cues of n u m b e r , position, a n d figure.
T h e first division of the plant k i n g d o m - - into those with a
lignaceous habit a n d those with a h e r b a c e o u s o n e - - derives f r o m
p o p u l a r classification; however, it can also be justified o n the
A r i s t o t e l i a n a s s u m p t i o n that the first f u n c t i o n of the "vegetative
soul" is n u t r i t i o n , a n d that the basic difference in habit reflects a
f u n d a m e n t a l difference in the m o d e of acquiring n u t r i m e n t . 37 I n
the s e c o n d place:
there m u s t b e a "vegetative" s u b s t a n c e allowing plants to r e p r o duce. This is the m o s t i m p o r t a n t aspect w h e r e v e r it is perfected.
T o this e n d were created the fruits a n d parts assuring the
fructification. This particularity does n o t exist a m o n g all plants,
only in the m o s t elaborated. A n d as a f u n c t i o n of the similarities
a n d differences in the fructification, o n e must establish secondary g e n e r a for the trees as well as for the m o s t h u m b l e plants.
It matters little w h e t h e r these g e n e r a b e n a m e d or not; for they
have n o t all b e e n n a m e d . O n l y those of e v i d e n t utility for m a n
have b e e n n a m e d , such as the legumes a n d grains.
Now, some ancient naturalists, such as Pliny, did partly attempt to establish
natural affinities via "sympathy" in superficial resemblances; but others, like
Aristotle, did not. Too, some Renaissance naturalists, like Brunfels, adhered to
the "doctrine of signatures"; however, Cordus avoided it and Cesalpino explicitly
rejected it. To be sure, the analogies employed by Aristotle and Cesalpino were
just as much free creations of the mind as symbolic speculations. Actually, their
analogies contained terms that also figured in symbolic metaphors of the time
(e.g., the plant as an upturned animal). Yet, what matters for science is not where
speculative ideas come from, but how they are subsequently treated. They
realized that contrary to mystical analogy, scientific analogy must be cautious
(Theophrastus, Historia plantarum l, i, 3--5; Cesalpino 1583:1) and must aim
ultimately to reduce itself to "dead metaphor," not to produce eternally openended "truth." Symbolism, in other words, is neither a cognitively necessary nor a
historically apparent stage in the passage from common sense to science. As for
the "colored and diverse fact," it already appears somewhat bleached in Brunfels
and Weiditz and positively monotone in Cesalpino.
37. For Cesalpino, roots and shoots were the parts primarily responsible for
fulfilling the nutritive function. The root would absorb the food from the soil,
while the leafy shoot would assimilate the absorbed food and distribute it to the
other parts of the plant. Accordingly, the group of woody plants is characterized
by the fact that the root and shoot are stronger and harder (habitiori substantia, &
duriori constant) than in herbaceous plants (1583:27). Note, however, that Ray
(1686:51--53) effects the same primary division by the presence or absence of
buds during winter, rather than by the hardness of the medullary substance.
236
scott
ATRAN
A n d if there were need of a third "vegetative substance," it

would be necessary to institute a third partition - - and to study
in the same way this element as well as the parts destined for it
a partition that would divide the preceding genera. But
insofar as the functioning of plants is s u m m e d up in the two
elements previously discussed, the classification and division of
genera will only take into a c c o u n t those elements. N u m e r o u s
genera rightly a p p e a r to follow the m o d e of fructification; for in
-
no other part has nature produced such a multitude and distinction of organs as are observed in the fruit. (Cesalpino 1583:27)
Thus, Cesalpino's further divisions of the two main groups are
based on those parts in which the next most important function of
the plant is located, namely, the fruits and seeds. 38 T h e characters
derived f r o m these parts are primarily meant to d e m a r c a t e taxa at
what is roughly the level of the m o d e r n family on up to the folk
life-form. 39 Presumably, the fructification characters are e n o u g h to
delineate these taxa because the information content of the fructifi-
cation is sufficiently detailed to make reference to other parts

merely redundant. So, when the fructification is constant, differences in parts of the stem, leaves, or roots are "in a sense
accidental" (1583:29). But this does not m e a n that other parts of
the plant can be ignored. True, "all the organs over and above the
root and pith (or 'heart') are for the p u r p o s e of fruit bearing"
(1583:6); however, "to the extent that they are for the sake of the
fructification" C q u a t e n u s fructus gratia datae sunt" [1583:9]), a
close attention to differences in these other parts may reveal
hitherto-unexpected differences in the fructification. M o r e o v e r ,
divisions below the family level m a y be provisionally m a d e on the
basis of parts other than the fructification; the herbaceous Leguminosae, for example, are subdivided according to possession or
38. Tournefort (1694:17) acknowledges that Gesner originally suggested the

importance of the seed, fruit, and flower for judging the relations between plants
in two letters written towards the end of his life; but it was undoubtedly
Cesalpino who first systematically applied such ideas. Yet it is not true that either
Gesner (Meyer 1854--1857, 4:334) or Cesalpino (Greene 1983, 2:808) had a
clear notion of a unique and uniform generic rank.
39. Cesalpino deals with seedless plants in bk. XVI, following his treatment
of the herbaceous seed plants (the existence of seedless woody plants was not
recognized at the time). On the whole, his seedless groupings reflect sound
morphological judgment: first come ferns and their allies (affines), next lichens
(lumped with heptatics), then mosses (with club-mosses and tree-mosses), algae,
and, finally, fungi. He also includes some seed plants that are not readily
perceived to germinate.
237
lack of tendrils. In the end, though, all secondary divisions down to

the level of species should proceed according to differences in the
fructification.
Reliance on the information value of the fructification is not a
direct deduction from a prior principles It derives from a painstaking attempt to delimit unequivocally an exhaustive series of
family-level "genera" in place of the intuitive and vaguely bounded
family-level fragments of c o m m o n sense. Because an exhaustive
series of species could no longer be expected to serve as a
complete basis for apprehending the order of the living world,
another basis would have to be found in accordance with readily
perceptible (commonsensical) notions of morphological affinity.
The partial local series of family-level fragments seemed to
provide the intuitive groundwork for just such a basis: 4 'q'he
plants are almost innumerable. So one must note intermediate
genera, hardly mentioned, and containing the most distant species;
but hitherto only a few have been revealed. For if one accepts the
grains and legumes, which are commonly called cereals and
vegetables, others can hardly be distinguished so clearly. And it
is especially use that in fact justifies this selection rather than
similarity in form -- which is what we are looking for" (1583:25).
Although Cesalpino rightly notes that ordinary folk rarely name
covert fragments, he errs in suggesting that they recognize only
those having some special utility. In fact, popular groupings that
depend largely on affinity in overall morphological aspect and that
"tend to be rather tightly focused on a single [biological] family"
are recognized in folk societies, though seldom named (Berlin
1982b; cf. Berlin, Boster, and O'Neill 1981; Bulmer 1979:62). It is
to such groups that Cesalpino adverts when he speaks of genera
innominata, the groups that (in the zoological realm) Aristotle
refers to as eide a n o n y m a . Such groups are not usually named
because, unlike the named folkbiological groupings of generic40. In Historiarum mundi (k. XX), Pliny divides the edible "grains" (fruges)
into cereals (frumenta) and legumes (legumina). But more often, higher-level
groupings are only implicitlyformed by listing kinds in sequence, as in the case of
the cucurbits. In like manner, Dioscorides implicitly takes stock of the labiates
simply by sequencing examples in his Materia medica (bk. III, chaps. 31--47):
lavender, origanum, thyme, pennyroyal, dittany, Marrubium, Ballota, Salvia,
Mentha, Satureja, and Marjoram; another sequence (chaps. 60--71) deals with
the umbellifers: Echinophora, Bupleurum, Angelica, Tordylium, Sison, anise,
caraway, dill, cumin, Amni, and coriander. Theophrastus is occasionally more
explicit in labeling, for example, the common "naked seed" Mediterranean
umbellifers anethon, koriannon, kuminon, marathon (fennel), and the like as
narthakados ( narthaks ~ ferula) (cf. Historia plantarum I, xi, 2).
238
SCOTT ATRAN
speciemes and life-forms, family-level fragments do not appear to

comprise a proper taxonomic series. That is to say, they do not
constitute a mutually exclusive, exhaustive, and relational partitioning of the local flora and fauna. For the herbalists as well, such
groups represent more or less isolated sections of weakly-linked
chains of generic-speciemes. The longer the section -- that is,
the greater the number of links -- the fuzzier the section's
composition. This is because any local environment is unevenly
riddled with morphological gaps at the "family" level. Nevertheless, whether the family fragment is named (cf. Taylor 1978-1979:227--228) or not (cf. Hays 1976:500), its morphological
composition is occasionally such that its members cross-cut plant
life-forms that do form a proper taxonomic series. Because overt
inclusion of such fragmenta would destroy any (transitive) taxonomic order in which plant life-forms also appear, they tend to
remain "covert" (Atran 1983).
This is likely the foremost factor in Cesalpino's decision to
sustain the separation, for example, between ligneous and herbaceous Leguminosae. For, on Cesalpino's own account, except
for the difference in stem habit, the two main classes of Leguminosae (i.e., woody and herbaceous) have the same differentiae.
His insistence on maintaining the distinction, despite an obvious
awareness of their morphological relationship, clearly derives from
his commitment to the intuitively elegant structure of popular
taxonomy, and not from blind adherence to metaphysical doctrine.
Cesalpino's dilemma is that of having to cross a conceptual
threshold beyond which common sense is no longer pertinent in
order to recapture the intuitive surety of common sense. The full
complement of nondimensional species could no longer be
grasped in its entirety; yet the nondimensional species is intuitively
the most secure contact that human beings have with the diversity
of the living world. Family-level fragments can provide only
intermittent intuitive contact with the local flora and fauna; but if a
way could be found, without violating life-forms, to complete an
unequivocal series of such fragments by evening out the local gaps
with knowledge obtained from the partial series of other environments, this would resecure the equivalent of local contact while
vastly extending it.
Determining such a series would require a calculated blend of
empirical and rational intuition. Like the herbalists, Cesalpino
must first secure all the intuitive knowledge thus far accumulated
relating ancient to local types: "as far as possible, I have felt it
incumbent upon me to match each separate plant to its ancient
names . . . but without neglecting popular appellations" (1583
239
dedicatory epistle). By taking known European types and fragments as standard referents, more remote environments could
then be scanned for species to fill out the fragmentary European
chains; but once completed, a full complement of European-based
family-level "genera" would presumably suffice as a summary
framework in which any exotic species might find its "natural
genus." If a foreign species were to occasionally show itself to be
misplaced, this would be owing to some insufficiency in the initial
description of the plant, and not to an insufficiency in the stock of
available genera:
In truth it is obligatory that in such a summary of plants, one
amongst the rest may accidentally escape us; as with a soldier
who sometimes sees himself placed in a division which is not his
own, so it may happen that a plant is classed in a genus which is
foreign to it. This especially concerns the medicinal plants of
far-off countries, of which only the root or sap or wood or
another part is transported, without us having the opportunity
to have ever seen the entire plant. (Cesalpino 1583, dedicatory
epistle)
Empirical intuition alone, however, could never ensure an
exhaustive and unequivocal delimitation of family-level "genera."

The European core of fragments would itself be grounded in an
intuitive appreciation of greater or lesser degrees of overall morphological affinity; foreign species would be attached as complements to that core on the basis of still weaker intuitions of affinity.
Thus, for indefinitely many exotic species, and even for morphologically isolated European species, no clear and unambiguous
decision on generic (i.e., familial) affiliation could be justified
intuitively. Consequently, the closure and well-boundedness of
individual genera as well as the completeness of the whole
sequence of genera would have to depend upon rational principles. Accordingly, all empirically intuited genera must, in some
way, be rigidly bound to a logical structure of fructification
characters. This is the task to which books II through XV of De
plantis are devoted. These books aspire to offer rational principles
of discovery, but they generally succeed only as easy-to-follow
keys to the identification of what was already intuitively known.
The way Cesalpino worked out his classification is speculatively
reconstructed by Bremekamp (1953; cf. Vines 1913), who suggests that he may have begun with a single intuitively wellgrounded group of plants or with several such groups. Cesalpino
likely chose the first track, argues Bremekamp, for that would
240
SCOTT ATRAN
h a v e led to the " d i s c o v e r y o f a large n u m b e r o f n a t u r a l g r o u p s , a n d

this . . . w o u l d h a v e e n c o u r a g e d h i m to p r o c e e d . ''41 But the
families a n d s e c t i o n s o f n a t u r a l families that C e s a l p i n o d i d p r o d u c e were, to a significant extent, a l r e a d y k n o w n to local folk a n d
the herbalists o f the time: G r a m i n e a e , L e g u m i n o s a e , Liliaceae,
Labiatae, Verbenaceae, Umbelliferae, Cruciferae, Ranunculaceae,
Cucurbitaceae, Euphorbiaceae, Compositae, Rosaceae, Caryophyllaceae, a n d P r i m u l a c e a e .
It s e e m s m o r e plausible, then, that C e s a l p i n o a t t e m p t e d b y trial
a n d e r r o r to find logically e l e g a n t c o m b i n a t i o n s of fructification
c h a r a c t e r s that w o u l d preserve these intuitive groupings. O c c a sionally, the logical p l a y o f such c h a r a c t e r s c o u l d i n d e e d l e a d to
the d i s c o v e r y o f n a t u r a l g r o u p i n g s that intuition of h a b i t u s a l o n e
w o u l d n o t likely reveal: for e x a m p l e f a c t o r i n g out the B o r a g i n a c e a e f r o m the L a b i a t a e b y the ovule's o r i e n t a t i o n . But the keying
p r o c e s s c o u l d also yield m o r p h o l o g i c a l l y m i x e d results: for e x a m ple, h e r b a c e o u s p l a n t s with b i l o c u l a r fruits d i v i d e into t h o s e with
a d o u b l e c a r y o p s i s (umbellifers), t h o s e with a single f r e e d s e e d in
e a c h fruit cell ( Mercurialis, Agrimonia, Poterium, Rubia, Galium),
t h o s e having several s e e d s in e a c h fruit cell with the d i s s e p i m e n t
p e r p e n d i c u l a r to the p l a n e o f s y m m e t r y (crucifers), a n d t h o s e
having s e v e r a l s e e d s in e a c h fruit cell with the d i s s e p i m e n t in
the p l a n e o f s y m m e t r y ( S c r o p h u l a r i a c e a e , p a r t o f S o l a n a c e a e ,
Plantago, Pirola, Potamogeton). A l l in all, t h o u g h C e s a l p i n o
m a n a g e d to p r o d u c e a logical k e y w h o s e c o n s e c u t i v e divisions
i n c l u d e p l a n t s that usually b e l o n g to o n l y o n e o r two n a t u r a l
families. 42
C e s a l p i n o clearly favors c e r t a i n a s p e c t s o f the fructification,
41. According to Bremekamp: "Cesalpino's starting point will have been one
of the strikingly uniform groups such as the Umbelliferae, the herbaceous
Leguminosae, the Liliifloraeor the Compositae. Let us suppose that he started
with the Liliiflorae.... He will have noted that the representatives of this group
are herbaceous plants provided with 3-1ocular fruits . . . as the presence of
3-1ocular fruits proved a useful character for diagnosing.., three groups, he will
have turned his attention towards plants that correspond with each other by the
presence of another number of fruit cells.., he will have noted that in the case of
plants with 2-1ocular fruit cells an even more natural arrangement could be
obtained by dividing them according to the same set of characters which had
proved useful for the subdivision of the plants with 3-1ocular fruits, viz., the
presence of one or more than one seed in each fruit cell. The subdivision led in
plants with 2-1ocular fruits to the splitting off of such well-defined groups as the
Umbelliferae and the Cruciferae. This success will have encouraged him to
proceed in the same way" (1953:585).
42. Not nearly so "deplorable" (Guyrnot 1941:19) an affront to intuitions of
natural affinity as might be inferred from cursory examination.
241
such as number of seeds per fruit, relative position of the petals

and stamens with respect to the fruit, and figure of the seed, seedreceptacle, and flower. But he provides no formal principle of
priority by which to automatically demarcate plant groups. N o r
does he fall back on functional priority. This lack of a definite
principle of priority no doubt owes to his realization that even if
the more obvious intuitive groupings may be tagged with fructification characters, the tagging sequence can apparently follow no
single logical line, or fundamentum fructifications, if such groups
are to be preserved. Yet without a single principled line of reasoning, the boundaries of the less-intuitive groupings would remain a
subject of indefinite controversy, with arbitrary claims being made
for the priority of different characters in different cases. 43
Later systematists sought to rectify the situation by removing
the problem of classification even further from neo-Aristotelian
concerns. The rational order would no longer be considered the
logical nexus where corruptible matter meets eternal godly form.
Rather, logical hylomorphism would cede to mathematical mechanism; that is, the rational order would be viewed as a mathematical trajectory tracing the workings of the Universal Artificer
through the living world. The goal would no longer be a formally
elegant indexing of plant forms, but a computational system whose
basic rationalization need only be evidence of a mathematically
efficacious sequence of morphological combinations. Such a system, Cesalpino could not envisage - - for he lacked the technical
vocabulary that would allow (Jung and Ray) a mathematical
description of plant parts, and he failed to fix the genus (in the
manner of Tournefort and Linnaeus) as a privileged level of reality
to which a principle of computation might be applied.
To know the logical place of any one form would be simultaneously to know the possible morphological similarities and dif43. Cesalpino's logical key is thus not quite a deduction. It takes the general
form of a dichotomous division, although some stages of the division have more
than two alternatives. At each step, a particular part (or complex of parts) of the
fructification presents at least two features that oppose one another in number,
position, or figure. But these opposing features by no means exhaust the logically
possible alternatives. Moreover, sometimes number is chosen as the criterion of
opposition, other times form or position; and occasionally an opposing feature
simply marks the absence of a certain number, position, or figure. In brief, no
attempt is made to exhaust all possible positive combinations of essential
characters according to some universal rule. No logical derivation of plant groups
is really offered: there is only a practical guide that unequivocally factors out
subordinate genera that are otherwise too difficult to recognize immediately,
either because they are too numerous or because their respective (overall
morphological)aspects tend to shade into one another.
242
SCOTT ATRAN
ferences between it and all other forms. For classification to

become this preset combinatory system of indefinitely many
morphological possibilities, the character of number would have to
shift from the ordinal to the cardinal; position would have to
translate into topological disposition in respect of structural morphology; and figure would have to transform into geometrical
configuration. In this way, a quantitative expression and reduction
might be given to multidimensional similarities and differences in
overall morphological structure. 44 Only then could one hope to
traverse, automatically and simply, the whole complex of morphological arrangements in the living world without having to perform
the superhuman feat of accumulating and committing to memory
all the morphological facts. The system would still have to be
made compatible with common-sense intuitions, though, for otherwise it could lay no claim to a vital link with the sensible world.
VI
For the better part of a century following the publication of De
plantis there was little sustained effort at systematic classification.
For the most part, Cesalpino's arrangement was ignored, although
his plant denominations were often cited, as herbalists gave
priority to describing the new species that exploration was revealing at an ever-increasing rate. Caspar Bauhin, for instance, knew
of Cesalpino's work, but he did not understand it and thought
that it would only confuse students of botany. The one major
exception in this period was Joachim Jung (Jungius), professor
of mathematics at Giessen and later of natural sciences at the
Akademisches Gymnasium in Hamburg. Educated at Padua, Jung
was undoubtedly influenced by both Cesalpino and Galileo. 45 He
sought to develop Cesalpino's natural system by creatively apply44. Division would proceed through each level by positive and opposed
characters that logically exhaust the conceptual possibilities available at each
level. Each level would be composed as a complete series of equal fractions of
the integral whole. Each progressively lower level would be composed of an
integral series of smaller equal fractions. But the smaller the fraction, that is, the
lower the rank of the taxon, the more intricate and complex the conformation of
variables that constitute the fraction. Thus, all fractions would have a constant
numerator (i.e., one or unity), but the equal fractions of each lower l~vel would
have a common denominator proportional in complexity to the mathematical
conformation of fructification characters defining the taxa of that rank: the lower
the level, the more complexthe conformation.
45. From the catalog of works in Jung's library complied by Christoph
Meinel (1984b) of the Universit~it Hamburg, it appears that Jung was conversant
with nearly all the published works of Galileo and Cesalpino. Concerning
243
ing the m a x i m o f G a l i l e o ' s H Saggiatore ( 1 6 2 3 ) that the b o o k o f

n a t u r e "is w r i t t e n in m a t h e m a t i c a l l a n g u a g e a n d the c h a r a c t e r s a r e
triangles, circles, a n d o t h e r g e o m e t r i c a l figures."
Jung's a p p r o a c h to scientific m e t h o d o l o g y , h o w e v e r , differs
f r o m b o t h C e s a l p i n o a n d G a l i l e o in that it d o e s not use d e d u c t i v e
analysis - - w h e t h e r logical division o r g e o m e t r i c a l d e m o n s t r a t i o n
- - e i t h e r as a t e c h n i q u e for the s o l u t i o n o f specific e m p i r i c a l
p r o b l e m s o r as a m e a n s to the a p p r e h e n s i o n o f self-evident
m e t a p h y s i c a l principles. L i k e H a r v e y , w h o was also e d u c a t e d at
P a d u a , Jung rejects t h e c e n t r a l t e n e t o f logical methodus e x p r e s s e d
in the w o r k s o f the g r e a t P a d u a n A r i s t o t e l i a n , G i a c o m o Z a b a r e l l a .
A c c o r d i n g to Z a b a r e l l a ( 1 6 0 8 , cols. 1 3 3 - - 1 3 4 ) , logic is the instrum e n t "by w h i c h w e a r e h e l p e d in a c q u i r i n g k n o w l e d g e of things";
h o w e v e r , "acquisition" is m e a n t in two ways, o n l y the first of which
w o u l d p r o v e a c c e p t a b l e to Jung a n d H a r v e y : as a set o f teaching
p r o c e d u r e s (ordines), logic alms to c o m m u n i c a t e existing k n o w l e d g e w i t h o u t ambiguity; b u t as a m e t h o d o f d i s c o v e r y (methodus
proprie dicta), it a s p i r e s to p r o v i d e n e w k n o w l e d g e (cols. 1 3 8 - 139), b e i n g an i n s t r u m e n t for y i e l d i n g " k n o w l e d g e o f w h a t is
u n k n o w n f r o m k n o w l e d g e of w h a t is k n o w n " (cols. 2 2 6 - - 2 3 1 ) . 46
F o r Jung a n d H a r v e y , though, the r e s o l u t i o n o f p a r t i c u l a r scientific
puzzles, as well as the d i s c o v e r y o f the p r e m i s e s o f any scientific
(or metaphysical) demonstration, are obtained by induction from
experience.
Cesalpino, in particular, Dr. Meinel informs me that while Jung attacked much of
"scholastic" Aristotelianism he was generally sympathetic to, and even inspired
by, Cesalpino and Zabarella. Cesalpino is most evident in Jung's botanical
thinking, but also in the composition of the Praelectiones physicae: "In fact the
authors most frequently quoted by Jungius were Zabarella (24% of all quotations), Aristotle (16%, more than half of which comes from Meteor.), and
Cesalpino (5% [most extensively from De metallicis, Niirnberg 1602]), followed
by Galen (2%)" (personal communication).
46. Ever since Cassirer (1906:134--41) claimed that Galileo's notions of
scientific method and explanation had much in common with Zabarella's, a
debate has ensued as to whether indeed modern scientific theorizing was initially
a further development of the Aristotelian exegetical schools of northern Italy
(Randall 1940) or, rather, a complete breakaway from them (Skulsky 1968).
Certainly both Galileo and Zabarella relied on deductive reasoning as a guide to
necessary physical truth -- a view that other "enlightened" Aristotelians, such as
Harvey and Jung, rejected, as did such "modern" exponents of inductivism as
Bacon and Newton. But Gilbert (1963) convincingly argues that Galileo's method
is more directly inspired by Euclid's account of geometrical analysis, which
Galileo converts into a model for inquiry in natural science. At best, contends
Jardine (1976), Zabarella's use of syllogism and regressus appears to serve as a
stimulating foil in the Galilean dialogues, whereby the reader is enticed into
agreement with the superior merits of Galileo's "geometrical" method.
244
SCOTT ATRAN
Although Jung and Harvey were well acquainted with Bacon's

work, their view of induction remained decidedly Aristotelian:
either by increasingly refined observation (Jung) or by repeated
experience (Harvey), empirical simples would presumably manifest themselves, as would their combination into necessary truths.
From this standpoint, a noble task of scientific investigation would
be to prove previous theories -- including one's own -- false by
rendering observation more exact. But while Harvey (1981:443-453) relies on analogy for extrapolation and argumentation from
immediate sense data, Jung (1982) looks to geometry.
Jung conceives of the scientific method as a critique, or
doxoscopus, that inextricably combines logic and experience. It is
geometry that provides everyday observations with the precision
needed to make the scientific connections between them manifest:
Just as the doctor must first purge the body of those substances
that make it sick, so the task of doxoscopus must be to prepare
a tabula rasa of the intellect in which the things of nature may
be allowed to project themselves. Thus for Jung, the scientific
study of nature consists exclusively in the observation and
description of those primary experiences in a nature initially
undefined, then differentiated and defined. Already Galen had
furnished the only sure method: start out from the most evident
and general givens of everyday experience in order to progress
little by little towards more difficult observations . . . . It is only
after having found and defined the lowest levels of being that
one could then constitute an axiomatic science of nature, in the
manner of geometry that starts from the point, the line, the
circle, or the parallels. (Meinel 1985; 1984a)
The only explicit methodological principle that Jung accepts is
Occam's razor; for this justifies the geometrical analysis of experience whose goal is to achieve a complete synthesis and resolution
(resolutio) of complex experience by a finite combination of
elementary principles. These principles are not, as with Galileo,
mathematical principles of discovery that allow the natural scientist
to abstract from the secondary qualities of sensible things so as to
apprehend "real" underlying physical forms. Rather, they are
strictly principles of sense experience, whose mathematical expression must be constructed after the fact. 47
47. In all of this, Jung's approach was destined to be a minor, if valiant, side
effort in the theoretical development of natural philosophy. But in natural history
there was a more sustained and fruitful wedding of enlightened Aristotelian and
245
Whereas for Galileo, Euclid's geometry provides a (partial)

model of reality, for Jung the structure of the "geometrical" model
in chemistry, as in natural history, is something that cannot be
prejudged in advance of empirical scrutiny. There could be no
mental anticipations of empirical generalization. Nor could one
allow contrived experiments to confirm or deny such a priori
considerations. So, there could be no conception of "law of
nature" or "scientific method" in Galileo's modern sense. But if
mathematical inference and law do not constitute the syntax and
semantics of Jung's book of nature, at least geometry provides its
alphabet.
Unpublished in his lifetime, Jung's botanical lectures were
edited shortly after his death by two former students, Martin Fogel
and Johannes Vagetius. Through the five fragments of the De
plantis doxoscopiae physicae minores (1747 [1662]), Jung proposes that all plants can be "accurately reduced to true genera by
specific differences, according to the rules of logic" (69). In the
first fragment, he questions whether the traditional life-form
divisions of tree and herb are valid. Citing Theophrastus's description of the otherwise herbaceous mallow occasionally becoming
"tree-like" (74), he argues that this is but one example where the
popular distinction appears to break down. 48 Such a division
cannot, therefore, be admitted as "philosophically" proper. He
emphasizes the importance of the geometrical configuration and
the topological interrelations of the floral parts and fruit for
delimiting genera, but he also stresses the significance of leaf structure. Like Cesalpino, Jung acknowledges that variations in spiny
modern scientific ideas; and this is understandable, given that the Aristotelian
cosmos was built on a biological model, not on a model of inert substances. It is
at this historical juncture that Jung's ideas of analysis found a special niche,
but one that was methodological rather than epistemological. Both empirically
and rationally inclined natural historians could profitably make use of his
mathematized descriptions of plants without committing themselves to Jung's
Aristotelian conception of science as classification and definition.
48. For Theophrastus (Historia plantarurn I, iii, 2), mallow, when it grows
tall, departs from its essential nature (physis) much as we might say that a bonsai,
although still essentially a tree, can "look like" something else (e.g., a shrub, or
even an herb). Jung, however, does not think that the philosophical difference
between intrinsic "nature" and mere "appearance" applies to such cases, most
likely because he realizes the irrelevance of botanical life-forms to a natural order
that is based wholly on morphology and is independent of the (local) economy of
nature. But the systematic relevance of life-forms, though again questioned by
Magnol (1689) and Rivinus (1690), was not finally decided until Linnaeus.
246
SCOTT ATRAN
texture, color, odor, taste, medicinal properties, location, and time

of germination are nonessential (77); but unlike Cesalpino, he
argues that the number of flowers and fruits are accidental as well.
The Doxoscopiae had little direct impact on the course of
systematics, but the short lsagoge phytoscopica (1747 [1679])
established the foundations of analytic plant morphology. A
handwritten manuscript was given to John Ray in the late 1650s,
and its influence owes chiefly to Ray's use of its morphological
principles in his own later systematic work. In the Isagoge the root,
stem, leaf, flower, and fruit are decomposed into analytically
significant elements. The leaf, for example, is characterized as that
which is laterally extended with respect to its point of attachment,
such that there is a difference between its external (adaxial) and
internal (abaxial) surfaces. The form of the leaf may be simple
(e.g., lily), concave (e.g., onion), or compound. Compound leaves
can be digitate (e.g., clover), paripinnate (e.g., pea), imparipinnate
(e.g., rose), or triangulate (e.g., celery). The leaf margin is either
whole or segmented, and if dissected, then laciniate, serrate,
crenate, or dentate. Even more important for the development of
systematics is Jung's distinction of the parts of the flower: calyx,
corolla, stamens, style, and ovary (fruit). Some flowers have no
calyx (e.g., tulips and lilies), but when there is one it is either
undivided (perianthium simplex) or divided (perianthium compositum). He sees that the number of lobes of the corolla marks
off different species, and that flowers with five or six lobes occur
more often than those with three or four. Although, like Cesalpino,
he ignores the reproductive role of the stamens, he associates
differences in their number, disposition, and configuration with
different species; his description of the fruit, however, does not
improve on Cesalpino's.
The restoration of interest in Cesalpino's system owes its chief
impetus to a curious debate between the leading English naturalists of the second half of the seventeenth century. In his Praeludia
botanica (1669:469) the king's physician and botanist, Robert
Morison, attacked the unnamed author of the tabular classification
of plants presented in John Wilkins's Essay Towards a Real
Character and Artificial Language (1668). At the request of
Wilkins, John Ray -- the first eminent naturalist who was not a
physician -- had drawn up these tables in conformity to Wilkins's
ideal grammar, proposing various alternative ways for dividing the
same groups and including such "accidental" features as taste,
color, odor, texture, and location. In a letter dated May 7, 1669,
Ray acknowledged that he knew the tables to be "manifestly
imperfect and ridiculous," for they basically represented a lin-
247
guistic r a t h e r t h a n a n a t u r a l d e m o n s t r a t i o n ( 1 8 4 8 : 4 1 - - 4 2 ) . N e v e r theless, he was c o m p e l l e d b y M o r i s o n ' s clear i m p l i c a t i o n that the

a u t h o r o f the t a b l e s was b o t a n i c a l l y i n c o m p e t e n t to p r o d u c e a
m o r e c o h e r e n t system (cf. C. R a v e n 1942).
M o r i s o n p r o p o s e d his o w n system against the "hallucinations"
o f the b r o t h e r s B a u h i n a n d c l a i m e d it to b e d r a w n solely f r o m
nature; h o w e v e r , his a s s e r t i o n that the nota generi'ca a r e to b e
sought exclusively in the fructification e v i d e n t l y d e r i v e s f r o m
C e s a l p i n o . 49 R a y h a d e a r l i e r p r o p o s e d a n a l p h a b e t i c a l c a t a l o g o f
the local C a m b r i d g e flora as a p r e l u d e to a c o m p l e t e British flora;
b u t at t h e c o n c l u s i o n to the c a t a l o g ' s Index he imagines, p l a u s i b l y
with Jung's r e c e n t l y c o m m u n i c a t e d m a n u s c r i p t in mind, that "it
w o u l d b e p o s s i b l e to classify p l a n t s in m a n y o t h e r ways, for
e x a m p l e in r e s p e c t o f their r o o t s o r stems o r flowers o r seeds,
t h o u g h it is n o t m y i n t e n t i o n to p u r s u e the subject f u r t h e r " ( R a y
1 6 6 0 : 1 0 3 ) . F o l l o w i n g M o r i s o n ' s attack, R a y set o u t to a p p l y Jung's
analytic m o r p h o l o g y to C e s a l p i n o ' s system, thus a b s o r b i n g all that
was w o r t h w h i l e in M o r i s o n ' s m e t h o d while e n d o w i n g it with a
p r e c i s i o n that w o u l d e n s u r e m u c h g r e a t e r success.
In the p r e f a c e to his M e t h o d u s p l a n t a r u m nova ( 1 6 8 2 ) , R a y
argues that classification s h o u l d n o t follow utilitarian o r e c o l o g i c a l
criteria; rather, in line with C e s a l p i n o , it s h o u l d b e b a s e d on "the
similitude a n d likeness o f the p r i n c i p a l parts." L i k e C e s a l p i n o , he
implicitly allows c o n s i d e r a t i o n o f all the e l e m e n t s o f the fructification, a l t h o u g h he e m p h a s i z e s "principally" the s e e d s a n d s e e d r e c e p t a c l e s . O n o c c a s i o n he also a c c e p t s as " c h a r a c t e r i s t i c " m a r k s
49. According to C. Raven, "the discovery of Morison's copy of Cesalpino

heavily annotated in his own had gives the lie to his explicit repudiation of
indebtedness" (1942:186). Tournefort was already well aware of Morison's
apparent plagiarism: "One would not have praised this author enough, except that
he has praised himself a little too much; far from being content with the glory of
having executed a part of the most beautiful project ever made in botany, he
dares compare his discoveries to those of Christopher Columbus . . . . One might
have taken him at his word, were it not for his having transcribed entire pages
from Cesalpino and Columna . . . . Ray, without making as much noise, succeeded
infinitely better than Morison" (1694:18).
Apart from his revival of Cesalpino's method, Morison is perhaps best
remembered for the Plantarum umbelliferarum distributio nova (1672), much of
which is incorporated in sec. IX of his Plantarum historiae universalis oxoniensis
(1680--1699). The umbellifers had been called by name since the chapter De
umbelliferis herbis in Dodoens's Stirpium historiae (1583), but Morison's treatise
amounts to the first real monograph of a botanical family. Unfortunately, the
quality of the fourteen other sections of Morison's Historia often falls far short of
sec. IX, except where modified by Bobart (in the posthumously edited 1699
volume) to conform more closely with Ray.
248
SCOTT ATRAN
drawn f r o m the f o r m of the roots or the locations of the leaves on

the stem (e.g., Asperifoliae [i.e., the labiates]). Like Jung, R a y
c o n c e d e s that the division of plants into life-forms is "popular"
rather than "philosophical," yet he chooses to retain the traditional
distinction anyway (24--25) His most significant d e p a r t u r e f r o m
Cesalpino, however, c o n c e r n s the division immediately following
that of life-forms: the division of herbs into those with either two
seed-leaves (dicotyledons) or only one ( m o n o c o t y l e d o n s ) would
eventually b e c o m e a standard item in the classification of flowering plants. 5
In the Historia plantarum (1686), R a y further elaborates his
system in a c c o r d a n c e with Jung's m o r p h o l o g y , and fixes the
criterion of conspecificity in the m a n n e r of Cesalpino:
differences that issue f r o m the same seed, be it in an isolated
plant or in an entire species, are accidental and not the signs of
a specific character . . . . T h e same is true of the animal world
the n u m b e r of species in nature is certain and determined:
God rested on the sixth day, interrupting his great work -- that
is, the creation of new species N o w the n u m b e r o f plants
varying by the color of their flowers or their multiplicity is
infinite: each year new variations are spawned which we reject
and to which we refuse the title and the dignity of true species
.. these variations are due to changes in the sky, the earth, or
the alimentation. (Ray 1686:40) 5~
These works mark a steady i m p r o v e m e n t in the delimitation of

natural groups, especially those subdivisions that c o r r e s p o n d to
m o d e r n families of phanerogams. Yet, like Cesalpino, R a y fails to
apply a single consistent basis for division t h r o u g h o u t his classification in o r d e r not to violate the m o r e readily perceived intuitive groups. His "synoptic" tables resemble Cesalpino's logical key
m u c h m o r e than they do a strictly principled division. In contrast,
50. Although not utilized as a basis for classification, there is implicit
recognition of the distinction between monocots and dicots among the ancients as
well as among tribal folk. Theophrastus, for instance, is well aware of the distinction between, e.g., monocotyledonous legumes and dicotyledonous cereals, and
the Tenejapa Tzeltal lexically generalize the distinction between monocots and
dicots (Berlin, Breedlove, and Raven 1974:63).
51. As for ostensible cases of hybridization or "transmutation": "Observe,
however, that such transmutation exists only between related species of the same
genus; and it is likely that certain of these are not characterized by specific
differences" (Ray 1686:43). This is hardly proof that Ray, as opposed to
Linnaeus, held species to be inconstant and variable, as Singer (1959:194)
suggests.
249
the Leipzig botanist, physician, and chemist Rivinus (Bachmann)

would shortly propose an elegant system whose major divisions
depend solely on structural variations in the corolla of the flower.
The definitive form of Ray's reply to Rivinus's Introductio generalis
in rem herbariam (1690) appears in an exchange of letters reprinted in the appendix to the second edition of Ray's Synopsis
methodica stirpium britanicarum (1696a). Here Rivinus argues
against the time-honored division of plants into life-forms because
it would not be fitted into a system employing "but one (sort of)
natural character" (Ray 1696a:4). As for Ray's own classification,
Rivinus remarks:
You concede the utility of characters chosen from the flower
and fruit in order to distinguish subalternate genera. I feel the
same, but while I consistently use them you waver, abandoning these essential marks whenever it seems appropriate. (Ray
1696a:21)
To this, Ray replies that a classification that conforms to nature
must occasionally use many characters drawn from the whole
habitus. Indeed, no single part is truly essential:
The correct and philosophical division of any genus is by
essential differences. But the essences of things are unknown to
us. Thus, in place of these essential characters, characteristic
accidents should be u s e d . . . [that] join together plants that are
similar, and agree in primary parts, or in total external aspect,
and which separate those that differ in these respects. (Ray
1696a:30--33)
At this juncture Ray's argument already has a definite Lockean
flavor. But as Sloan (1972) notes, it is only in formulating a reply
to the much more formidable system presented in Tournefort's
t~ldmens de botanique (1694) that Ray clearly embraces some of
the main ideas of Locke's Essay 1848 [1689]):
The essences of things are wholly unknown to us. Since all our
knowledge derives from sensation, we know nothing of things
which are outside us except through the power that they have to
affect our senses in some particular way, and by the mediation
of these impressions to cause a particular image to arise in the
intellect. If the essences of things are immaterial forms, it is
admitted by everyone that these are not encountered in any
sensible means. If they are really nothing other than certain
250
SCOTT ATRAN
fixed proportions and mixtures of elements or natural minima,

then these minima cannot be perceived by our senses, except
with several instruments or aids. (Ray 1696b:5)
Tournefort's classification displays even more logical elegance
than Rivinus's, while doing less violence to naturally perceived
affinities. Nevertheless, argues Ray, the fungi have neither fruit nor
flower, and most deciduous plants lack these organs for much of
their lives. How, then, can such nonexistent or transient characters
be "essential" to the life of plants? Moreover, since the roots and
stem are always present, these organs would have a better claim to
being essential, although hardly anyone would consider using only
such parts as a basis for classification.
In his last important work, Methodus plantarum emendata et
aucta (1703), Ray concedes that the fructification is usually
preferable to other parts; however, it is by no means an infallible
guide to the natural order of genera. In the section "De methodo
plantarum in genere" he writes:
For the memory and the right method one must, by way of a
sign of recognition, adopt either one property with a generic
character or a mark which is c o m m o n to all the species already
known . . . . This mark is very readily gotten from the flower or
the fruit of those plants which exhibit these marks, because
these parts vary less often in the same genus than do the others;
and they are, by heir color, figure, and other accidents, conspicuous and easily observable. Nevertheless, if in the future
one encounters a plant which agrees in other parts and accidents with the rest of the species of the genus, but does not have
the mark which we have taken for characteristic, then such a
plant must be included in the genus with which it has several
attributes in common.
The fructification is the most "natural" character for classification
only in the sense that, of all the parts, it has the highest information content. All things being equal, it is usually redundant with the
whole habitus. The natural priority of the fructification thus owes
not to its functional essentiality, but to its practicality. In other
words, it best fulfils the tenets of Occam's razor: "The ctraracteristic marks of the genera must not be multiplied beyond necessity,
nor accumulated to an extent more than necessary to determine
the genus with certainty."
The fructification, though usually sufficient, is not always necessary. A proper characterization of any given genus must thus
Origin o f the Species and G e n u s C o n c e p t s
251
include a description of the entire habitus as a necessary preliminary. O n l y then can a characteristic m a r k be c h o s e n which is
r e d u n d a n t with the whole habitus, be it a m a r k selected f r o m the
fructification o r f r o m s o m e other part:
A c o m p l e t e definition is m a d e up out of a genus proximum and
an essential difference: but the essences of things are u n k n o w n
to us, h e n c e the essential differences of them also. H o w e v e r ,
since f r o m the essences flow the same qualities, operations, and
other things which are accidents, there can be no surer m a r k o f
essential, and so of generic, agreement than to have m a n y parts
and accidents similar, o r to have the whole facies (faciem),
habitus (habitum), and contexture (texturam) the same. (Ray
1703)
A l t h o u g h R a y follows L o c k e in claiming the real essences of
beings to be unknowable, and the c o m p l e x perceptions arising in
o u r minds to be only nominal essences, he differs f r o m L o c k e on
two crucial points. First, R a y never a b a n d o n s his belief in the
absolute reality of species:
I would not d e n y that universals have a f o u n d a t i o n in things
truly agreeing or similar in special parts or properties. This
agreement is so great, especially in living things, that individuals
of the same species are seen as having been m a d e according to
the same exemplar or idea in the Divine Mind . . . . F r o m this it
follows that species are distinguished essentially f r o m one
a n o t h e r and are not transmutable, and the forms and essences
of these are either certain specific principles, that is, certain
very small particles of matter, distinct f r o m all others, and
naturally indivisible, or certain specific seminal reasons enclosed by means of an a p p r o p r i a t e vehicle. (Ray 16 96b:vi)
A t the level of the species, then, the nominal essence unequivocally signals the presence o f a real essence, even though the
underlying principles of the real essence m a y remain forever
obscure. 52
52. In his Essay, Locke sees matters quite differently: "I would n o t . . , deny
that nature in the production of things, makes several of them alike: there is
nothing more obvious, especially in the races of animals, and all things propagated by seed. But y e t . . , two species may be one, as rationally as two different
essences be the essence of one species; and I demand, what are the alterations
that may, or may not, be in a horse or lead, without making either of them to be
25 2
SCOTT ATRAN
S e c o n d , R a y n e v e r s e e m s to h a v e a c c e p t e d L o c k e ' s s k e p t i c a l
c o n c l u s i o n that a true s c i e n c e o f n o m i n a l e s s e n c e s is i m p o s s i b l e .
F o r Ray, n o t o n l y species, b u t h i g h e r g e n e r a as well, c a n b e k n o w n
to b e real. F o r L o c k e , h o w e v e r , claims as to the reality of a b s t r a c t
types, w h e t h e r species o r g e n e r a , a r e "wholly unintelligible, a n d
w h e r e o f we have s c a r c e so m u c h as any o b s c u r e o r c o n f u s e d
c o n c e p t i o n in g e n e r a l " (Essay III, vi, 10). Such claims, argues
L o c k e , a r e the c u l t u r e - b o u n d illusions o f o r d i n a r y language. 53 Ray,
b y contrast, c o n s i d e r s p h e n o m e n a l intuitions to b e s t r o n g i n d i c a tions o f true genera: so m u c h so, in fact, that he a p p e a r s to
a b a n d o n e a r l i e r h e s t i t a t i o n s c o n c e r n i n g the " p h i l o s o p h i c a l verity"
o f life-forms. I n d e e d , o n R a y ' s o w n a c c o u n t i n g m o r p h o l o g i c a l
intuitions a r e s u r e r signs o f n a t u r a l a g r e e m e n t t h a n any p h i l o s o p h i c a l principles.
In a c k n o w l e d g i n g the u n i v e r s a l validity of habitual, u n s k i l l e d
intuitions b u t o n l y the p r a c t i c a l i t y o f fructification c h a r a c t e r s , R a y
has n o m e a n s b y w h i c h to settle definitively t h e p h e n o m e n a l
b o u n d a r i e s o f c o m m o n - s e n s e g r o u p i n g s within a w e l l - f o r m e d
n a t u r a l o r d e r . A n o p t i m i s t i c e m p i r i c i s t might see n o p r o b l e m here,
b u t for a s k e p t i c o r a rationalist this w o u l d n e v e r do. E i t h e r the
quest for a n a t u r a l s y s t e m b r e e d s only t e m p o r a r y illusions o f
a g r e e m e n t , o r a p r i o r i c o n s t r a i n t s o n a l l o w a b l e possibilities m u s t
b e r i g o r o u s l y a d h e r e d to. In effect, T o u r n e f o r t ' s rationalist a l t e r n a tive s i m p l y inverts R a y ' s priorities, a d m i t t i n g c o m m o n - s e n s e intui-
another species? In determining the species of things by our abstract ideas, this is
easy to resolve; but if any one will regulate himself herein, by supposed real
essences, he will I suppose be at a loss; and he will never be able to know when
any thing precisely ceases to be of the species of horse or lead" (III, iii, 13).
53. Only individuals truly exist. Nature itself is indeterminate, as the
existence of borderline cases proves. We choose to ignore such cases only out of
a linguistic habit for "presumptive ideas of several species" (Locke, EssayIII, vi,
29) -- ideas that "receive their birth and signification from ignorant and illiterate
people, who sorted and denominated things by those sensible qualities they found
in them" (III, vi, 25). As it is for common sense, so it is for more self-conscious
systems that merely extend "this gibberish, which, in the weakness of human
understanding, serves so well to palliate man's ignorance" (Ill, x, 14). If discovery
and exploration have shown anything at all, it is that so-called "natural kinds" are
strictly relative to human interests and purposes. What is constant and singular
for one linguistic community is variable and plural for another. Only the scientific
community's concerted search for a common vocabulary, requiring "much time,
pains, and skill, strict inquiry and long examination," can reasonably hope to
produce those complex associations of simple perceptions which most probably
reflect species "conformable to those of nature." Such is the task of natural
history (III, xi, 24), which may yield proper philosophical sorts wholly different
from anything that common sense might initially lead us to suspect.
25 3
tions as useful but making the fructification characters universally

d e c i s i v e , a t l e a s t i n p r i n c i p l e (cf. T o u r n e f o r t 1 6 9 7 ) . 54
VII
D e s c a r t e s i n s i n u a t e s i n Discours de la m d t h o d e t h a t t o t r a n scend the bounds of sense one must substitute analytic certainty
f o r s e n s i b l e i n t u i t i o n : 'q~he a n a l y s i s o f t h e A n c i e n t s . . . is s o
c o n s t r a i n e d b y a c o n s i d e r a t i o n o f f i g u r e s t h a t it c a n n o t e x e r c i s e
the understanding without greatly tiring the imagination" (1907
54. The argument between Ray and Tournefort is not, as Sloan (1972)
suggests, between progressive Lockean empiricism, on the one hand, and regressive Aristotelian essentialism on the other -- for Tournefort is no more an
Aristotelian essentialist than is Ray. Rather, the debate pits Ray's mildly
optimistic form of empiricism against Tournefort's mildly skeptical form of
rationalism: if Ray follows Locke in believing real essences to be unknowable, he
does not follow Locke in believing that nominal essences can never directly
reflect reality; and if Tournefort follows Descartes in thinking that the true order
of things may be indicated by a priori principles, he does not follow Descartes in
thinking that such true indications necessarily reveal anything of real essences.
Nor is Tournefort any less committed to atomism and mechanism that is Ray;
Gassendi's theory of "molrcules," for instance, is very much in evidence in
Tournefort's medical thesis of 1695 (cf. Bianchi 1957) and in subsequent lessons
given at the Jardin du Roi in Paris (Tournefort 1708).
Sloan makes much of the fact that Boyle (and, by implication, Locke and Ray)
rejects the Aristotelian distinction between essence and accident in favor of the
distinction between primary and secondary qualities. Whereas both essence and
accident are thought to objectively inhere in a thing, and both are presumably
knowable through their direct phenomenal manifestations, for Boyle (1667) only
purely subjective secondary qualities are knowable -- that is, the colors, tastes,
textures, sizes, and shapes of objects conveyed to us by the senses. The primary
qualities that determine the "real" form of the object refer to the insensible
motion, mass, and geometrical arrangement of the minute corpuscular constituents of the object. Because primary qualities are insensible, they are unknowable. Yet, there are other versions of the primary-secondary distinction that do
not entail the claim that we can never know the real order of things. In II
Saggiatore, for example, Galileo enumerates the primary qualities of matter: these
are either geometrical (shape, size, position, contiguity), arithmetical (number), or
kinematic (motion or rest). The senses provide us also with secondary qualities of
color, taste, texture, and so forth; but it is reason alone that allows us to know
that there are primary qualifies, and to know what, where, and when they are.
Galileo advances this distinction to justify his opposition to Aristotle's ban on the
use of mathematics in natural science. Clearly, the "essential" characters of
Tournefort (and even Linnaeus) are more in conformity with Galileo and
Descartes than with Aristotle. For knowledge of nature must be consonant with
quantitatively expressible, clear, and distinct ideas that reason ties to what is
readily observable.
254
sco'Iq" ATRAN
[1637], pt. 2). Similarly, for Tournefort part of the rationale for
botanical system is that "the study of plants does not greatly tire
the imagination when undertaken with method" (1694:4).
To ensure the system a firm grounding in reality, Tournefort is
obliged to establish a basis where the conflict between reason and
customary modes of apprehension is minimal. This privileged
point of contact between nature and art is the genus. Before
Tournefort, "genus" referred to any group expressly including two
or more species. With Tournefort, "genus" comes to denote only
that conceptually fixed and perceptually unitary level of reality
which logically ranks immediately above that of the species, but
which is otherwise perceptually confluent with the species, that is,
with a bouquet of morphologically similar individuals (1694:13).
Now, intuitively given genera are usually associated with distinct fructifications. Thus, on the basis of the rationalist principle
of sufficient reason, there must be a naturally unique indexing of
known and as-yet-unknown genera by fructification features. If
gaps should appear in the sequence of characters, one must not
forget that God has a horror of vacuums. This ensures, a priori,
that genera still to be discovered by expedition will already have a
fixed place within a preestablished series, whether or not we are in
fact able to reason out what the series is. Implicit in this approach
is the Cartesian credo of veracitas dei: if God has undeceivingly
allowed us to see a part of nature's plan, then he has given us
reason by which to anticipate the whole.
As Tournefort puts it:
the author of nature has fixed the appropriate mark~ of plants.

. . . Now if the genera are confounded, how do we represent
species to ourselves? Nature, so it is said, seems wholly alien to
any method. But even if this be conceded, experience convinces
us that without a method in things so numerous and diverse as
in botany one could but comprehend nature obscurely. Those
who imagine that the correct distribution and naming of species
depend upon men's fancy are not less mistaken. The Creator of
all things, who gave us the faculty for giving names to plants,
places in the plants themselves signifying marks from which
should be sought that similarity required of species of the same
genus. We can neither change these marks nor renounce examining and using them, if we would eliminate error. (Tournefort
1719:54)
If we fail to uncover the whole of nature's plan by such clear and
255
distinct characters, it is because we, and not G o o d and nature, are

imperfect. 55
At first glance, Tournefort's idea of the "naturalness" of genera
seems to refer only to what is convenient for the mind: "It is much
more convenient to reduce most of the known plants to 600
genera than to reduce them merely to two or three hundred,
because most of these two or three hundred will be overloaded
(charge) with so many species that one will not be able to express
their differences except by highly composed names" (1694:38).
But it is clear from his insistence that genera must be "true"
that Tournefort means by "naturalness" more than mere ease of
memory. He advises that species judged to be "heteroclites or
bastards or degenerates, if such terms be permitted," should not
be allowed to encumber well-defined genera "because they do
not have the essential marks of those genera to which they are
attached"; in such cases, the heteroclites should be allowed to form
distinct genera of their own, even if the genus contains only one
species: "What is the necessity, for example, of following Morison
in calling hops, Convolvulus heteroclitus perenni, floribus foliaceis,
strobili instar? Wouldn't it be better to make of it a particular
genus, and to leave to it the name, Lupulus vulgaris, which is
known the world over?" (1694:38).
Tournefort argues that the ostensibly mixed generic character
of a heteroclite is an indication of a truly distinctive generic
discontinuity in nature, which further exploration and more careful
observation may confirm with new species. Nonetheless, the possibility exists that a genus will remain monospecific (e.g., the
Belladonna); and it is this possibility that underscores the fact that
the genus represents a fixed level of reality, no matter what its
specific content.
The crucial question arises, though, why six hundred genera?
Certainly the names of plants can be reduced to a mediocre
number, if one wishes to fix on those which are necessary. One
will have, so to speak, the key of this science in retaining the
names of about 600 genera, under which can be brought the
greatest part of known plants. It would be useless to burden the
m e m o r y with all the synonyms that one has given them . . . . If
55. As Descartes himself notes, "our ideas not notions, being real things
which come from God insofar as they are clear and distinct, cannot to that extent
fail to be true. Accordingly, if we often have ideas which contain falsity, they can
only be ideas which contain some confusion and obscurity.., that is to say, they
are confused in us only because we are not whollyperfect" (1907 [1637], pt. 4).
256
SCOTT ATRAN
after this cutting back (retranchement) one still thinks of complaining that the names of plants are too numerous, it would
be to accuse nature of being too fecund in its productions.
(Tournefort 1694:3)
The key to understanding the naturalness of genera, then, is the
notion of "cutting back," or retrenchment, of the 6,000 to 10,000
known species of plants by one order of magnitude to the same
number of basic-level groups entertained by the ancients, by the

German, Italian, and Dutch herbalists, and by other folk (cf.
Raven, Berlin, and Breedlove 1971). In short, the genus emerged
as the "natural" accommodation of the rational mind to a perceptual reality that could no longer be contained within the
ordinary bounds of common sense.
The unequivocal nature of the fructification would leave no
doubt as to generic character. If the habitus, rather than the
fructification, were chosen as the basis for the genus, however,
then the genus would be bound to a morphological type. The
choice of generic habitus would thus depend upon which species
was chosen as the type. Such a choice would invariably be the
subject of disagreement among authors who, because of their
familiarity with their own environments, might prefer one species
over another. The extension of the generic grouping would vary
with the type chosen, and there would be potentially as many
genera as species. In that case, the endeavor to replace limited
common-sense surety with an equally certain global order would
be doomed to failure, and the quest for a natural system would
have to be abandoned.
Still, the need to reduce habitual structure to a characteristic
part does not suffice to explain why the fructification was chosen,
rather than the stem, leaf, root, or some combination of these
parts. The traditional account has it that, by identifying the genus
with its fructification -- that is, with its reproductive parts -- the
genus could also assume the metaphysical status of a natural,
generative force in the world. T o classify by fructification would
then be tantamount to explaining how the natural order necessarily came to be (cf. Sachs 1890 [1875]; Sloan 1972).
Tournefort, however, explicitly denies that functional importance has any bearing on the choice of the fructification or certain
of its elements as essential generic characters: "Let one not say that
nature, having no goal other than the production of fruits, requires
us to consider these as the most noble parts of plants. There is no
question here of nature's intentions, nor of the nobility of parts;
rather, it is a question of finding expedients for the sake of
257
distinguishing plants with all possible clarity; and if the least of

their parts were m o r e appropriate than those which are called the
most noble, they should be preferred" (1694:26). Actually, he goes
on to argue, the roots appear to be the parts most essential to the
life of plants, but they are scarcely indicative of the natural order
of plants. So it would seem that Ray's criticism of Tournefort's
reliance on "essential" parts that are, in fact, not immediately
crucial to the life of plants is beside the point.
Tournefort's "definition" of the plant in terms of root, seed,
leaves, stem, and flower is a direct appeal to lay knowledge: "All
men have m o r e or less the same idea, and [these parts] would be
rendered m o r e obscure, if one were to define them" (1694:21). 56
In this respect, he is simply rendering explicit what herbalists and
systematists before him had implicitly assumed to be the "natural
parts" of plants. T h e true nature of these parts, which any real
definition would have to describe, could not be specified so long
as the real nature of things were unknown to us. It suffices that we
know what is, and what is not, a natural part.
Thus, starting with what is presumably both a natural and a
customary decomposition of plants into their constituent parts,
Tournefort proceeds to examine those elements that might be
taken from the fructification as against those elements that might
be taken from the other parts. H e concludes that elements taken
from other parts are too numerous if treated together; that is, "it is
dangerous to descend into too great a detail, for fear that plants
discovered in the future may lack a few of the essential parts
thought to belong to the character of their genus" (1694:24). Or,
they are too few if taken separately; for example, "differences of
roots and stems are too small in number" (25). In the one case,
there could be an unsystematic influx of new genera; and in the
other, there might not be enough genera recognized for discerning
an ordered connection between them.
The fructification would never have been adopted as the basis
for botanical classification were it not for the fact that it usually
56. Evidence from ethnobotany seems to confirm the cross-cultural aspect of
this decomposition. Friedberg (1972), for instance, indicates the importance of
the names that the Bunaq, for Timor give to these parts (cf. Greene 1983, 1:137
for the ancient Greeks). Similarly, for the Tzeltal Maya, Berlin et al note: "We
have arranged that plant part names recognized in Tenejapa Tzeltal into groups
of expressions referring to stem, bark, leaves, roots, flowers, fruits, and so on . . . .
Their recognition of the relevant aspects of plant morphology compares
favorably with that of trained Western systematic botanists, given the restriction
that they worked unaided by microscopes" (Berlin, Breedlove, and Raven
1974:69, 79). Also unremarkable in this respect is the ancient Chinese protoencyclopedia, E R H Y A (contrary to claims by Needham 1986:126-- 135).
258
SCOTT ATRAN
accords with intuitions about which bouquets of plants seem

naturally to go together. That is to say, the fructification is largely
redundant with the whole habitus - - not because it is the most vital
part, but because it is the most complex morphologically. Still,
the fructification does not always correspond to intuition, as
acknowledged implicitly by Cesalpino and explicitly by Ray. A
successful development of the system, then, would turn on the
quest for a definitive resolution of the inevitable conflict between
systematic art and intuitive nature.
Tournefort concedes Ray's point that many plants, including
algae, mosses, and mushrooms, have no visible fructification
pattern at all; hence, they must be diagnosed by some other part,
until such time as a "true character" (vdritable caract~re) can be
found in them which would be analogous to the visible fructification. Even when such characters are manifest, however, they are
not always sufficient; for "if one only had regard for the flowers
and fruits, it would be necessary to change several things which
custom has, with some reason, authorized: for example, wheat, rye,
oats, barley, and all the species of couch-grass would have to
belong to the same genus, since their flowers and fruits have much
the same conformation" (1694:28--29). 57 But it would be wrong
to think that these grasses belong to the same genus: "One can thus
establish as a general maxim in botany, that the flower a~d the
fruit are necessary parts for establishing all genera whose species
manifest flowers and fruits, but these parts do not always suffice to
distinguish genera from one another" (30). In cases ,,vhere the
fructification alone does not suffice, "second-order" genera may be
established "in relation to other parts."
Moreover, Tournefort concedes that his, or any, natural system
can only be considered partial and tentative. F o r him, a "natural"
system of the rational mind only m o r e or less follows the doubtlessly perfect logical thread by which G o d weaves together the
world: "It is necessary to note that whatever method one uses,
however exact it may be, one will always believe it to be defective.
. . . I know, moreover, that all new rules are subject to many
contradictions, and I do not doubt that something first will be
found to be said aginst the system which I propose (1694:42). 58
57. Sloan wrongly claims that for Tournefort "the characters of the flower
and fruit are always necessary and sufficient for defining the genera of plants in
those cases where these characters are manifest" (1972:48).
58. According to Dughi, it is precisely through such considerations that
Tournefort manages to distance himself from his more scholastically minded
predecessors and successors: "The conception of the importance of organs of
259
Still, it w o u l d b e an e r r o r to c o n c l u d e , as A d a n s o n ( 1 7 6 4 : c v ) a n d
Buffon ( 1 7 4 9 , 1 : 1 8 - - 1 9 ) d o , that T o u r n e f o r t thus r e g a r d s his
classification as a n artifical k e y d e v o i d o f a n a t u r a l basis: " T o u r n e fort, n o t r e g a r d i n g his m e t h o d as natural, b u t as artificial, p l a c e d
his g e n e r a in the s a m e rank. L i n n a e u s t o o k his p r e t e n t i o n s
further." O n l y with r e s p e c t to h i g h e r - o r d e r classes can the
m n e m o n i c simplicity o f a single, visually a g r e e a b l e floral diagn o s t i c b e c o n s i s t e n t l y p r e f e r r e d to m o r e a p p a r e n t l y natural, b u t
vague, intuitions of o v e r a l l affinity - - a n d e v e n this p r e f e r e n c e m a y
b e justified o n l y to the e x t e n t that it l e a d s to an effective k e y i n g o u t o f n a t u r a l genera. T h a t is w h y h i g h e r - o r d e r g r o u p s s h o u l d b e
b a s e d o n c h a r a c t e r s d r a w n f r o m the s a m e p a r t o f the p l a n t that is
u s e d to establish t r u e genera.
In fact, T o u r n e f o r t j u d g e s the fructification to b e the o n l y
c o n c e i v a b l e n a t u r a l basis for a system. H e d o u b t s o n l y that his
basis is c o m p l e t e . T h i s d o u b t has a t w o f o l d c h a r a c t e r . First, t h e r e
is a w a r e n e s s o f the fact that e v e r y t h i n g that has to b e m a r k e d is
n o t yet fully a v a i l a b l e for analysis: " w h a t e v e r p a r t o n e takes, it is
sure that o n e will n e v e r find a m e t h o d w h i c h is a b s o l u t e l y general,
b e c a u s e . . . o n e d o e s n o t k n o w t h e flowers a n d the fruits o f all
the p l a n t s " ( 1 6 9 4 : 4 3 ) . 59 S e c o n d , e v e n if the flowers a n d fruits o f
reproduction taken from Cesalpino led most notably Linnaeus to an all-tooscholastic principle that 'natural' genera must be defined uniquely with the aid of
characters drawn from the five parts of the flower. . . . The choice of these floral
characters is no longer essentially inspired, as with Tournefort, by considering
their diagnostic value, nor corrected by the eventual admission of vegetative [i.e.,
habitual] characters, nor, most importantly, completed by figures; conceived a
priori it is restrictive by design" (1957:181). Dughi goes much too far, though, in
attempting to distinguish Tournefort's concern with rational diagnosis from the
"scholastic" concerns of Cesalpino and Linnaeus. Concern with rational diagnosis
is equally apparent in Cesalpino and Linnaeus, as is the willingness to admit
vegetative characters, at least provisionally. Furthermore, Linnaeus rarely, if ever,
diagnoses a plant (or animal) without first analyzing its entire figure. What
distinguishes Tournefort from Cesalpino and Linnaeus, then, is not adherence to
rationalist principles, or admittance of other than fructification characters in the
construction of the system; rather, it is the fact that Tournefort considers the
fructification characters to be epistemically, but not ontologically, essential to the
natural order of plants. In other words, Tournefort does not believe, as Linnaeus
does, that possible variations on the patterns of fructification represent the
potentially viable modifications on the theme of life itself; for reason cannot
discover nature's real intentions, only its resultant contours.
59. For cryptogams, Tournefort offers to mark genera provisionally by the
character of the facies. Although systematists anticipated analogies between the
fructification of flowering plants and the reproductive apparatus of the nonflowering plants, no lasting advance in the systematic treatment of cryptogams
occurred before the detailed study of bryophytes by the German physician
260
SCOTT ATRAN
all plants were readily discernible, current notions a b o u t which

characters are sufficient would invariably change with exploration
and discovery (43).
Linnaeus thought to resolve the d o u b t s inherent in T o u r n e f o r t ' s
system by d e c o m p o s i n g the fructification into a precise c o m b i n a tion of analytic elements; for only a demonstrative science of
b o t a n y based o n rational principles could be p r o v e d a true science
(1751, sec. 19): "what I like are distinctions that are definite
and capable of d e m o n s t r a t i o n " ( 1 7 3 7 b , sec. 267). Faith in the
epistemological (computational) rationale for the system would
once again be linked to the acceptance of a metaphysical (essential) rationale (1751, sec. 88). 60 E a c h genus would thus be associated with a characteristic formula; and mathematically regular
patterns would be sought a m o n g these characteristic formulae to
c o m p u t e all fundamentally possible basic (i.e., generic) states of
plant life: "These marks are to us as so m a n y plant letters, which, if
we can read, teach us the character of plants; they are written in
G o d ' s hand; it should be o u r task to read them" (1737a, Ratio
operis).
As a result, according to Linnaeus, discovery of new generic
forms could be anticipated that would a c c o m m o d a t e as-yetu n k n o w n species into the eternal o r d e r of things: "The system is
for b o t a n y the thread o f Ariadna, without which there is chaos. Let
one take, for example, an u n k n o w n plant of the Indies, and let a
botanophile leaf through descriptions, figures, every index; he will
not find it unless by chance. But a systematist will determine it
straightaway, be it old or new . . . . T h e system indicates the plants,
even those it does not mention; this, the e n u m e r a t i o n of a catalog can never do" (1751, sec. 156). T o this end, he divides the
flower into four parts (calyx, corolla, stamen, pistil) and the fruit
into three (pericarp, seed, receptacle) (secs. 86, 92). E a c h of the
Hedwig. From Ray's "Musci" to A.-L. Jussieu's "Acotylrdones," cryptogams

occupied the phenomenally "residual" place in the plant world that the invertebrates did in the animal world.
60. Like Locke, Linnaeus came to believe that real essences are effectively
unknowable. Yet he never doubted that reason gives a true indication of where to
find them. In line with Tournefort, he held that the character determines only the
possibility of knowing the genus, and does not itself constitute the essence of the
genus: "know that it is not the character which makes the genus, but the genus
which makes the character" (1751, sec. 169). But contrary to Tournefort, he
considered the character a true mark of the underlying essence of the plant. In
appealing to reason and the fructification, Linnaeus also differed from Ray, who
sought to mark the location of real essences by aspects of the whole habitus
primarily grounded in phenomenal intuitions, rather than in rational principles.
O r i g i n of the Species a n d G e n u s C o n c e p t s
261
seven parts of the fructification is f u r t h e r s u b d i v i d e d into as m a n y

e l e m e n t s as readil3~ strike the eye at first glance (seven for the
calyx, two for the corolla, three each for the s t a m e n s a n d pistils,
eight for the pericarp, a n d four each for the seed a n d receptacle).
L i n n a e u s t h e n calculates that these t h i r t y - o n e elements, each
c o m p r i s i n g the four d i m e n s i o n a l variables of n u m b e r , configuration, disposition, a n d p r o p o r t i o n (magnitude), would "suffice for
3 , 8 8 4 generic structures, o r m o r e t h a n will ever exist" (sec. 167). 6~
T h e s e generic structures r e p r e s e n t the u n e q u i v o c a l m o d e s of
possible existence, only s o m e of which are e v e n actually realized.
H a v i n g thus calculated all possible basic existential modalities,
he is therefore r e a s o n a b l y assured that the fructification w o u l d
p r o v i d e for a n y n e w discovery, i n a s m u c h as he has a l r e a d y
s u c c e e d e d in r e d u c i n g nearly 7 , 0 0 0 of the 1 0 , 0 0 0 or so species of
plants then estimated to exist into a few h u n d r e d genera.
T h u s did L i n n a e u s see himself as a " s e c o n d A d a m " able to
e n u m e r a t e all the world's basic kinds, just as the first m a n had
d o n e in Paradise. T h e p r e s e n t w o r l d c o u l d b e a p p r o a c h e d as the
ideal local e n v i r o n m e n t writ large: g e n e r a w o u l d o c c u p y the role in
the overall n a t u r a l e c o n o m y of the earth that each of the p r o t o typical species had o c c u p i e d in Paradise. 62 T h e totalizing aspect of
61. The first Stockholm edition of the Philosophia botanica puts the figure at
5,736, but this expresses an error both in the addition of parts and in the
multiplication of the sum total of parts by the four variables. The correct figure
appears in the fourth edition (1787).
62. In the Oratio de Telluris habitabilis incremento, Linnaeus reveals
Paradise to have been an ideal local environment in which all basic kinds were
harmoniously concentrated. Linnaeus's Paradise is conceived to have been an
island in the midst of waters that covered the rest of the earth. In fact, it is
modeled on Tournefort's (1717, 2:xix) first-hand description of Mt. Ararat made
during his journey to the Levant. Here are to be found all the world's climatic
zones, together with the distinctive flora and fauna representative of each zone:
"It is worth remembering what Tournefort relates in the tale of his voyage to the
Orient. Assuredly, he rediscovered by the base of Mt. Ararat plants he had
previously seen in Italy. Climbing a bit higher, he will have encountered plants
that grow near Paris. The plants characteristic of Sweden are in a place more
elevated; but those which are proper to the mountains of Switzerland and
Lapland occupy the most elevated place of the mountain very near the snowcovered summit" (Linnaeus 1972 [1744]:37). As the earth's landmass, which
originally consisted of Paradise alone, spread upon the waters, it did so in such a
way that the climatic zones of the primordial island-center simply expanded in
proportion to their initial dimensions. Hence, the present world may be divided
into rather homogeneous areas: tropical, alpine, temperate, etc. Although at first
Linnaeus thought that all the world's species had been represented in Paradise by
a founding couple (or hermaphrodite), he later (1790 [1760]) implies that only
the prototypical species of each genus was initially present.
262
SCOTT ATRAN
folkbiological knowledge, that had been lost since the Renaissance

now seemed on the verge of recovery for the world at large, and
Paradise regained.
CONCLUSION
F r o m Cesalpino to Ray, the family-level covert fragments of
E u r o p e a n folkbiology substantially underlay the genera proxima
(or genera majora) - - centerpost of systematic classification.
During that period systematics was very largely geared to completing the arrangement of E u r o p e a n species into larger intuitive
groupings. To the E u r o p e a n naturalist's eye, the sequence of these
larger groupings is pocketed with lacunae; that is, E u r o p e a n
family-level fragments just do not cover the whole of the E u r o p e a n
flora and fauna (cf. L a m a r c k 1785:442, 451). But implicit in the
early systematic work was the notion that one could round out the
series of E u r o p e a n fragments by attaching foreign species to them,
thereby simultaneously creating a framework for systematically
incorporating new forms without reducing them to the old. 63
As long as this assumption remained unchallenged the strategy
would endure. But the realization gradually dawned that the
handful of better-known E u r o p e a n fragments were too fuzzy at
their boundaries and too p o o r in content to readily or rigorously
a c c o m m o d a t e the ever-increasing variety of exotic species. The
completion of a worldwide system of family-level taxa would thus
have to await a knowledge of all basic exotic groupings. This left
the fundamental problem of reducing the world's species to
manageable proportions unresolved; for it was precisely the prior,
or "automatic," incorporation of basic foreign groupings that the
system sought to ensure. Such is the difficulty that Tournefort's
genus was designed to overcome. Henceforth, the quest for a
complete series of natural families (corresponding to Cesalpino's
"genera") would have to await a knowledge of the basic (Tournefortian) genera.
63. Although a prior intimation of family-level groupings may have encouraged the domestication of related plants, it is hardly plausible that domestication would have preceded family-level perceptions, as Li (1974:719) claims. In
fact, there is no evidence that the family-level groups commonly recognized by
folk are initially restricted to cultivated plants. Nor is it likely that the supposed
symbolic virtues of such plants as the rose first animated the perception of a
wider network of family resemblances (cf. Walters 1961), though symbolic or
domestic value may have put a premium on naming and firmly bounding certain
family-related groups.
263
A final step was required, however, before the principle of a

rationally complete "natural system" of (Tournefortian) genera as
a basis for a "natural method" of families (or ordines) could be
rightly formulated. This involved the abandonment of those folkbotanical life-forms (especially tree and herb, but also shrub,
undershrub, and vine) that continued to dominate the systems of
Cesalpino, Ray, and Tournefort; for such life-forms often violate
natural families and hence block a coherent sequencing of genera
within and between families. This step, though already intimated
by Jung, Rivinus, and Pierre Magnol, was only first properly
appreciated by Linnaeus.
It is not size or orientation relative to man that alone is
important in the appreciation of botanical life-forms, but place in
human ecology; that is, life-form divisions seem to be made on the
basis of those habits of life that determine the place of each being
in the local economy of nature pertaining to man's everyday life
(Atran 1985a). The ecological importance of life-forms is attested
to by the fact that modern botany retains them only in ecological
contexts. Nevertheless, this subjective appreciation of what plants
are relative to us has little bearing on an objective appreciation of
what plants are relative to one a n o t h e r once a local environment
has been transcended: trees are bigger than people, and grasses
(herbs) smaller; trees are where birds most often perch and where
quadrupeds forage; trees determine exposure to sun and moisture
and hence the density of other vegetation, and the possible
habitats of many of the animals familiar to local folk. Yet, such
totalizing ecological frameworks are scarcely of value in organizing
knowledge of the world of nature outside the orbit of man's
everyday life.
Even within a local community it is plain that certain groups of
plants possess striking affinities that either cross-cut life-forms
(Leguminosae, Rosaceae) or fall between them (Malvaceae). For
this reason Linnaeus definitively banned life-forms from taxonomy.
Although they may be locally valid products of our "natural
instinct" (Linnaeus 1751, sec. 153), when devoid of ecological
context they become "scabrous" and "lubricious" (sec. 209) if
utilized as frameworks for organizing flora worldwide. Freed from
the provincial straitjacket of life-forms, the search for a generic
order that could hold together family fragments and link them to
one another might thus be effectively pursued. 64
-
64. That there is a definite anthropocentric bias in these pretheoretical

divisions cannot be gainsaid. But, again, such cognitive bias is too often confounded with some socially elaborated, or utilitarian, viewpoint (cf. Croizat
264
SCOTT ATRAN
But n o s o o n e r h a d a r e a s o n e d f o r m u l a t i o n o f the s i m u l t a n e o u s
p r i n c i p l e s of a "natural s y s t e m " o f g e n e r a a n d a "natural m e t h o d "
o f families finally e m e r g e d , t h a n it was o v e r t h r o w n . T h e factors
r e s p o n s i b l e for the d e m i s e of the genus as the p r i n c i p l e r a n k in
t a x o n o m y c a n n o t b e e x p l o r e d h e r e w i t h o u t delving fully into the
origins o f the family a n d class c o n c e p t s - - a n d this, s p a c e d o e s n o t
allow. B e c a u s e this p a p e r is c o n c e r n e d chiefly with the origins o f
the species a n d genus c o n c e p t s , b r i e f m e n t i o n of the following
m u s t suffice (see A t r a n 1986).
A t its i n c e p t i o n , the defining c h a r a c t e r of the genus - - that is,
the fructification - - was crucially a r a t i o n a l n o t i o n , a l t h o u g h
m e t a p h y s i c a l l y s a n c t i o n e d as the seat o f life. It r e q u i r e d c o n c e p t u a l
isolation o f t h o s e analytically p r i z e d c h a r a c t e r s of the visible fruit
a n d f l o w e r that c o u l d b e a p o d i c t i c a l l y a r r a n g e d into a p r e s e t
c o m b i n a t o r y system. T h e d e t a c h a b i l i t y a n d r e d u c i b i l i t y of visible
p a r t s to c o m p u t a b l e c h a r a c t e r s was, h o w e v e r , p r i m a facie less
w a r r a n t e d in the c a s e o f animals. T h e p a r t s o f a n i m a l s i m m e d i a t e l y
lend t h e m s e l v e s to c o n s i d e r a t i o n as f u n c t i o n a l l y i n t e r j o i n e d o r g a n s
r a t h e r than as visibly j u x t a p o s e d features.
In a d d i t i o n , the c o n s e r v a t i o n o f a n i m a l life-forms f o r e s t a l l e d
a t t e m p t s to d i s s o l v e animal (and t h e r e f o r e u l t i m a t e l y plant) k i n d s
into o n e vast a n d s e a m l e s s table o f r a t i o n a l characters. P e r h a p s
b e c a u s e we o u r s e l v e s a r e v e r t e b r a t e s , o u r a p p r e c i a t i o n of verteb r a t e life-forms is n o t far r e m o v e d f r o m an o b j e c t i v e a p p r e c i a t i o n
of m o r p h o l o g i c a l affinities b e t w e e n v e r t e b r a t e s themselves. In any
event, special c o n c e r n for t h e s e life-forms implicitly r u l e d out a
w h o l e s a l e a p p r o a c h to animal o r g a n i z a t i o n f r o m the start. F r o m
the R e n a i s s a n c e o n w a r d s , the analysis of z o o l o g i c a l f o r m s p r o c e e d e d m o s t l y within the f r a m e w o r k of s e p a r a t e m o n o g r a p h s
treating distinct a n i m a l life-forms. But b e c a u s e t h e r e w e r e initially
1945); yet, "tree" is no more derivative from a functional preoccupation with

wood than "green" is derived from a cultural preoccupation with plants in
general. In this respect, it is worth mentioning that plant life-forms have all along
continued to serve as phenomenally convenient, and not entirely "unnatural,"
frameworks for organizing an understanding of local flora for professional
naturalists who otherwise reject life-forms as general botanical taxa. Thus,
Linnaeus reports exotic genera in terms of life-form status; for example, in the
Critica botanica (1737b, sec. 238) he introduces three new American genera:
Hernandia (arbor Americana), a tree, Plumeria (arbuscula Americana), a bush,
and Milleria (planta Americana), an herb. Similarly, Bartlett (1935; cf. Bartlett
1926), in his report on the flora of the Batak lands of North Sumatra, employs
the sorts of life-forms used by the Batak themselves in organizing the local flora:
(a) trees and shrubs, (b) herbs, (c) ferns, and (d) mosses. Many an ethnobotanist
and field naturalist does likewise.
265
many fewer animals to worry about than plants (wild animals

especially being harder to spot in nature, to relocate for study, and
to examine without destroying), the problem did not seem at first
compelling in face of the welter of plant forms that occupied the
attention of those naturalists who were seeking a global system.
Recall that the genus was introduced in an effort to cope with a
number of plant species approximately one order of magnitude
above that which ancient and Renaissance herbalists ordinarily
faced. But awareness of new forms began to increase at a geometrical rate. Within Linnaeus's own lifetime the family replaced
the genus as privileged rank of taxonomy when the quantity of
classifiable forms rose yet another order of magnitude. Indeed,
given the difficulty of zoologically isolating a generic character and
rank, naturalists soon endeavored to forgo the quest for a "natural
system" of genera and to focus on the "natural method" of families
as a direct and immediate goal of systematics (Adanson 1764;
Jussieu 1789).
Eventually, though, the family too would lose its privileged
status, once it was realized that the family's advantage lay in its
capacity to furnish the most visibly abstract means whereby
common sense might gain access to the nonphenomenal world of
biology. The newly discovered world of "invisible" biological
causes and processes would no longer be hostage to the natural
historian's criterion of manifest visibility. Henceforth families
would be construed as no more, nor less, "natural" than genera.
The scientific development of basic common-sense classificatory dispositions thus occurred first at the level of the species and
then at the genus and family levels in order to accommodate an
increase in the quantity of phenomenal information that the mind
had to process. But the quality of information, including evergreater knowledge of overlapping phenomenal relationships,
posed an additional problem. On the one hand, neat sortings into
well-delimited morphological groupings became more difficult. On
the other hand, overlapping phenomenal relationships provided
prima facie evidence that living kinds were biologically interlinked
and that inferences might be made concerning the general distribution of vital properties between them.
Once taxonomy was grounded in an understanding of underlying organic processes, the study of those processes would soon
emerge as the principal object of a new science little constrained
by the phenomenal predispositions of common sense. It was now
zoology's turn to profit from a marked increase in the quality of
information (Farber 1982a, 1982b), including biological (physiological, embryological) and historical (paleontological) information
266
SCOTt ATRAN
still largely unavailable to botany. No longer could data concerning phenomenal relationships so completely dominate other sorts
of hitherto fragmented information. Accordingly, it was concluded
first in zoology (Cuvier 1812; cf. Daudin 1926--1927, 2:91--109)
and then in botany (Candolle 1833; cf. Stevens 1984b:67) that
there were only a few separate plans of organization in the living
world. Such plans, or embranchements, could by no means be
considered phenomenal plans, because they would accord no
theoretical privilege to relations among or between phenomenally
compatible genera (folkgeneric-speciemes), families (covert flagmenta), or classes (life-forms).
Still, whether or not phenomenally compatible higher-order
taxa prove to be ontologically sound, they undoubtedly provide
cognitive, if not epistemic, access to nature; for when set in a rigid
taxonomic framework they constitute a means of both storing
information and generating it. The synoptic advantage of a ranked
classification for information storage and retrieval is considerable.
To take a rather typical example from Jaume St. -Hilaire's Exposition des families naturelles (1805), which represents a synthesis of
the work of Linnaeus, Adanson, A.-L. Jussieu, and Lamarck: the
chapter dealing with the Leguminosae occupies some fifty pages;
of these, approximately one page is devoted to a general description of the family and one page to exceptions, possible links with
other families, and a discussion of previous literature; each of 100
or so genera is then analyzed in nearly half a page; and each of the
250 or so species is characterized in a few lines -- that is, only
about a page and a half of description is necessary for all 250
species. Any modern monograph would show this summarizing
effect of taxonomic structuring (Wharburton 1967). Ever since
Cesalpino, this synoptic character of taxonomy has been a motivating force in systematics.
But more than intellectual satisfaction or relief is involved.
Taxonomy provides a blueprint for a detailed study of biological
processes directly related to evolution: ecologic, zoogeographic,
phyletic, genetic, etcetera (cf. Bock 1973). It does this not only be
summarizing information already given, but also by supplying a
framework for finding new information through the systematic
extension of observed relationships: properties generally observed
in two species are tentatively assumed to be present throughout
the smallest-ranked taxon to which those species both belong.
Although in this respect the higher Linnaean ranks favored by
common sense cannot be considered ontologically or epistemically
privileged over other ranks of modern taxonomy (e.g., phylum,
tribe, order), they continue to enjoy a measure of practical advan-
267
tage. Because they are psychologically convenient, they have

historically provided, and continue to provide, a ready cognitive
access to (a first approximation of) scientific taxonomy.
What, then, of species? How can the differences between
species be considered of a fundamental nature when they appear
to owe their origin, at least in part, only to an accident of observation -- namely, the extinction of groups of varieties that have
disappeared from view? "In short, we shall have to treat species in
the same manner as those naturalists treat genera, who admit that
genera are merely artificial combinations made for convenience.
This may not be a cheering prospect; but we shall at least be freed
from the vain search for the undiscovered and undiscoverable
essence of the term species" (Darwin 1883:426). Although the
essence of species can be dealt with in this way, this does not entail
that what we commonly take to be species -- namely, interbreeding morpho-geographic communities of organisms -- are artificial
combinations. On the contrary, as with Lamarck (1809:74--75),
species really exist as naturally bounded sympatric groups. For
there need be no antagonism between the fact that species evolve
and the fact that they so obviously cohere as groups of coexisting
individuals in a local flora and fauna: "the more distinct two
populations are in space and time, the more difficult it becomes to
test their species status in relation to each other, but the more
irrelevant biologically this also becomes" (Mayr 1969:27). Evolution, in other words, is not a continuous movement of populations
through an environmental sieve, but a passage punctuated with
local adaptations to specific ecologic and geographic regulations.
Indeed, Darwin's initial formulation of the theory of natural
selection turns on a solution to the problem of how nature is
capable of producing commonly acknowledged species from races.
For Darwin, only those races count as species that are "ancient
and perfectly adapted" (notebook E 72, mid-December 1838). As
a consequence of slow and gradual adaptations and constitutional
divergence under selection pressures, relatively permanent and
deeply hereditary changes are wrought over time. These profound
historical developments in the race manifest themselves in the
sympatric species readily apparent to every naturalist:
The argument for natural selection would fail, in Darwin's eyes,
if it failed to be a solution to the problem of the origin of
species, a problem whose formulation presupposed that the
term had its customary reference. For the argument is directed
to explaining the origins of those entities that are already being
counted by naturalists as species. As Darwin understood his
268
SCOTT ATRAN
problem situation, the argument had to be so directed. The very

norms of the public enterprise that is science required it to be
so. And in this he was to be followed most explicitly be
Wallace; particularly when Wallace opened his account of
Darwin's theory in his book Darwinism by explaining what was,
in 1859, meant by the term species and what was meant by
discovering the origin of species . . . . We have here the basis for
Darwin's realism concerning species . . . as taxa. (Hodge
1985)
T o be sure, the species criteria that races must fulfil in order to
count as real species were not simply those of the folk naturalist,
namely, being a relatively well-delimited morpho-geographic community: the criteria of genealogical connection, nonhybridization,
and interbreeding also had to be met. But these criteria, which
natural historians had developed in response to emerging theoretical demands on basic folkbiological notions, were themselves
generally inferred from apparent constancy of morphological
character and ecological niche. In this sense, all conceptions of
species taxa from Cesalpino to Darwin were implicitly required to
meet conditions of compatible common-sense realism.
Where Darwin and Lamarck differ from their predecessors,
then, is not in denying the existence of species taxa, but in refusing
to acknowledge that the ranking of races into species is unequivocal. The "best" of species do pretty much fit the bill. But the
criteria can be met only approximately, never absolutely, even in
principle. This means there can be no essential difference between
species and variety. Nor, therefore, can the species category, or
rank, be ontologically true. Truth conditions for the delimitation of
species taxa, however, may be given by a correct definition of the
species concept. The best candidate around, and the one most
compatible with the folkbiological conception, is the modern
definition of the biological species concept proposed by Ernst
Mayr: "A species is a reproductive community of populations
(reproductively isolated from others) that occupies a specific niche
in nature" (1982:273). 65
65. Mayr's definition, which is today widely accepted, was elaborated after a
long and careful reflection on the nature of the avifauna of Highland New Guinea
and on local native understanding of the fauna. More than fifty years ago, Mayr
found that basic folk groupings nearly always correspond exactly with the
naturalist's delimitation of phenomenally salient sympatric species. For more than
forty years, this discovery was virtually ignored by biologists and ethnobiologists
alike.
269
The ecologic formation of communities of more or less distinct

sympatric groups evolving over time is an image captured easily by
the ordinary mind's eye. Here is a picture that seems to be largely
true to life, at least for most vertebrates and flowering plants that
constitute the setting for humankind's wonted existence. That we
spontaneously eye such things more or less as they really are is not
so very surprising, when our everyday life depends on it; and our
own evolution surely must have geared our minds to making such
dependence viable. Of course, c o m m o n sense itself has no more
to contribute directly to a biological unriddling of "the species
problem" than it does to a causal resolution of the higher ranks;
but it does provide the problem a transtheoretical grounding.
The congruence of folkbiological species concepts across cultures, together with the practical concordance of historically
elaborated species concepts, suggests a course of both realism and
rationality in the development of science from its common-sense
origins. Competing schools of thought may bicker over the significance of meter readings, but without the meter there could be
no dialogue at all. If the progress of science is measured, at least in
part, in terms of success in solving technical puzzles, then the
Assuredly, paleontologists (el. Burma 1949) and botanists (cf. P. Raven 1976)
generally have considerably more difficulty in ascertaining whether a given
population constitutes a biological species. However, this state of affairs in itself
does not undermine the adequacy of the biological species concept as a definition
of the species concept; it is just that in these cases the biological species concept
is not much help in actually delimiting species taxa. Other criteria must be used
to infer that a given population actually constitutes a species taxon and fits the
definition of the species concept. Inference from a population, or sample thereof,
to the existence of a biological species may be circuitous without being circular.
The definitions's adequacy depends on there being a theoretical justification for
each step of the inference and on the possibility of there being some pointed
empirical confirmation along the route.
There is little warrant for Dwyer's contention that the ecological species
usually apprehended by tribal folk constitutes the only "objective species" and
that reproductive criteria merely reflect "attempts to rationalize the 'reality' of the
species in terms of the ideology of evolution" (1976a:433). Insofar as taxonomic
species are presumed to be causally linked, and inasmuch as scientific criteria of
objectivity depend upon causal linkage, then evolution is the source of taxonomic
species. It is not the mere effect of geographical barriers that is responsible for
niche specialization; rather, genetic isolating mechanisms are the prime evolutionary causes of such competitive exclusion. Admittedly, criteria of reproductive
isolation are specific to particular environments and do not apply uniformly to all
the individuals of a species population; however, this does not mean that such
criteria can ever be dispensed with. It simply implies that, if there are any
objective criteria at all, they can only be applied statistically and locally, not
deterministically or independently of context.
270
scott
ATRAN
o n g o i n g i n q u i r y into the "species p r o b l e m " surely c o u n t s as o n e of

the basic set of scientific puzzles b y which to m e a s u r e success.
This is o n e m o r e a s s u r a n c e that science, in its quest for n e w
horizons, n e e d n e v e r go f r o m o u r m u n d a n e world e m p t y o r blind.
In o u r intellectual a l i e n a t i o n from usual s u r r o u n d i n g s we, as
scientists, m a y well d e n y that o u r familiar setting is the only
subject, o r e v e n a p r o p e r subject, for ontological inquiry; however,
it should b e kept in m i n d that w i t h o u t the p e r s e v e r a n c e of s o m e
c u s t o m a r y knowledge, a m o r e l e a r n e d p u r s u i t of what there is
w o u l d gad into myth. F a r f r o m acting m e r e l y as an "epistemological obstacle" to science, or b e i n g c o n c e p t u a l l y i n c o m p a t i b l e
with it, c o m m o n sense has played, a n d c o n t i n u e s to play, a
substantial part in tying scientific discourses to o n e a n o t h e r a n d to
the world. 66
T o recapitulate: Basic-level folkbiological groupings have always
p r o v i d e d a n intuitive u n d e r p i n n i n g a n d empirical a p p r o x i m a t i o n
for the scientific species. N o m o r e theoretical "justification" n e e d
b e given for this s p o n t a n e o u s o p e r a t i o n of c o m m o n - s e n s e realism
than for color p e r c e p t i o n o r the a p p r e h e n s i o n of t h r e e - d i m e n s i o n a l
rigid bodies. N o r was a n y given by the likes of T h e o p h r a s t u s ,
Cesalpino, T o u r n e f o r t , Ray, or L i n n a e u s , w h e n e v e r there was
express a p p e a l to c o m m o n sense. T h e difference b e t w e e n folk a n d
science here is that o n e gives p r i m a c y to the ecological gap
b e t w e e n p o p u l a t i o n s , the o t h e r to the r e p r o d u c t i v e gap. M o s t
often, though, the o n e is c o v a r i a n t with the other. 67 W i t h the intro66. Of course, the fact that a given conceptual framework is well tried and
never discarded wholesale does not warrant the claim that it is, ipso facto, true
and natural: "Do not work with stable concepts . . . . Do not be sbduced into
thinking that you have at last found the correct description of the 'facts' when all
that has happened is that some new categories have been adopted to some older
forms of thought, which are so familiar that we take their outlines to be outlines
of the world itself" (Feyerabend 1970:22). So, the fact of the relative constancy
of taxonomy across changes in theoretical interpretation is not necessarily proof
of the reality of taxonomy, as Dobzhansky (1937:305) seems to suggest. On the
contrary, it may indicate an epistemologically lax proclivity to put new wine in
old clouded bottles (Stevens 1984a). But if the relatively steadfast historical
thread is considered not a direct line on truth, but a filiation of real puzzles to be
explained -- a means of stitching successive theories to one another and to the
world that appears to us -- then the undeniable progress of science in understanding nature makes rational sense, even if nature ultimately dissolves both the
stitching and the appearance.
67. Thus, it is not surprising to fmd that basic-level folk taxa may, in the case
of species relatively unfamiliar to the modern biologist, initially prove a more
correct guide to taxonomic status than does scientific classification; following an
ethnobiological analysis of the Karam (New Guinea) frog speciemes kosoj and
wyt, for instance, biologists came to agree that there were indeed two species
271
duction of reproductive criteria, any new species could readily be

given cross-community status: the most commonly perceived
features of local species would also be those that usually happened
to breed true. The status that underscored the permanence of
these basically visible forms over time also sanctioned the principle of their systematic comparison and the possibility of their
grouping within higher groups that would transcend the bounds of
local knowledge.
Together with the introduction of specific breeding criteria, the
emergence of the genus concept was initially motivated by historical difficulties that exploration had posed for common sense.
The genus was designed to allow the reduction of species by an
order of magnitude to equivalence classes whose number and
quality the mind could easily handle again. The place of a new
species in the natural order of genera would be accessible either
(1) in terms of empirical intuition (readily visible morphological
agreement with a E u r o p e a n representative or some other preferred type-species of the genus) or (2) by means by intellectual
intuition (analytic agreement with the generic fructification according to the number, topological disposition, geometrical configuration, and magnitude of its constituent elements), and it would
ultimately be commensurate with both (allowing a mathematical
reduction of the new species to its associated type by reason of
their c o m m o n fructification). As a result, the customary surety of
the folk naturalist might be rationally extended to a worldwide
scale.
Unfortunately, this first systematic attempt to recover the
intuitive certainty of that common-sense paradise lost in the age
following the Renaissance was short-lived: the disparity between
fructification and habitus eventually proved much greater than
anticipated; the computation underestimated the task of reduction
by at least an order of magnitude; and no obvious generic standard
could be found for animals. Although a method for forming prac-
(Hyle angiana and H. disrupta), rather than the one (H. becki) previously
thought (Bulmer and Tyler 1968:376). Basic folk kinds, however, are only
empirically approximate. Thus, herpetologists have shown that morphologically
dissimilar and geographically separated local populations can actually belong to
the same species (e.g., Holbrookia maculata) and, conversely, that geographically
coexisting and morphologically similar populations may constitute distinct
species (e.g., Rana chiricahuensis and R. urticularia) (cf. Cole 1984). Even
though folk may be, or may become, aware of reproductive ties that do not fully
accord with morpho-geographic distinctions, they may well persist in believing
that only morpho-geographic kinds are basic (cf. Dwyer 1976b).
27 2
SCOTT ATRAN
tical generic units did develop, the q u e s t i o n of w h e t h e r g e n e r a - or taxa of any o t h e r higher r a n k - - are of a real a n d f u n d a m e n t a l
n a t u r e is still moot.
Acknowledgments
This p a p e r was c o m p l e t e d u n d e r the Scholar's A w a r d P r o g r a m

in the H i s t o r y a n d P h i l o s o p h y of Science ( N S F G r a n t No. SES8 5 0 7 8 9 6 ) . I a m i n d e b t e d to P. F. Stevens a n d Phillip Sloan for a
careful r e a d i n g a n d criticism of an earlier version. M y thanks also
to Jacques Roger, M. J. S. Hodge, a n d E r n s t M a y r who, with
Phillip Sloan, constructively engaged some of the m a j o r p o i n t s of
m y a r g u m e n t at the I n t e r n a t i o n a l C o l l o q u i u m , "Histoire d u C o n cept d ' E s p ~ c e dans les Sciences de la Vie," organized by the
F o n d a t i o n Singer-Polignac in Paris, M a y 1985.
REFERENCES
Adanson, M. 1764. Families des plantes. 2 vols. Paris: Vincent.
Albertus Magnus. 1955. De natura et origine animae, de principiis motus
processivL quaestiones super de animalibus. Vol. XII of Opera omnia.:
Monasterri Westfalorum.
Aquinas, Th. 1946. Tractatus de unitate intellectus contra Averroistas. Rome:
Gregorian University.
1955. Somme th(ologique [Summa theologiae]. Translated by A.-M.
Rouget. 6 vols. Paris: Desclte.
Arber, A. 1950. The Natural Philosophy of Plant Form. Darien, Conn.: Hafner.
1953. "From Medieval Herbalism to the Birth of Modern Botany." In
Science, Medicine and History, ed. E. Underwood, pp. 317--336. Oxford:
Oxford University.
Aristotle. 1908--1931. The Complete Works of Aristotle. Edited by J. A. Smith
and W. D. Ross. 11 vols. Vol. I: Analytica priora, trans. A Jenkinson, and
Analytica posteriora, trans. G. Mure [1928]; vol. IV: Historia animalium,
trans. D. Thompson [1910]; vol. V: De partibus animalium, trans. W. Ogle
[1912[; vol. VIII: Metaphysica, trans. W. Ross [1928].
-1963. Categories [Categoriae] and De interpretatione. Translated by J.
Ackrill. Oxford: Clarendon.
1975. On the soul [De animaI. Translated by W. Hett. London: Heinemann.
-1980. The physics [physica1. 2 vols. Translated by P. Wicksteed and F.
Cornford. London: Heinemann.
Atran, S. 1983. "Covert Fragmenta and the Origins of the Botanical Family."
Man, 18:51--71.
1985a. "The Nature of Folkbotanical Life-forms." Amer. Anthrop.
87:298--315.
1985b. "Pre-theoretical Aspects of Aristotelian Definition and Classification of Animals: The Case for Common Sense." Stud. Hist. PhiL Sci.,
16:113--163.
-1986. Fondements de l'historie naturelle. Pour une anthropologie de la
science. Brussells: Complexe.
-

- -
273
1987. "Folkbiological Universals as Common Sense." In Noam Chomsky:

Consensus and Controversy, ed. S. Modgil and C. Modgil, pp. 247--268.
London: Falmer.
Bachelard, G. 1980. La formation de l'esprit scientifique. Paris: Vrin.
Balme, D. 1980. "Aristotle's Biology Was Not Essentialist." Arch. Geschichte
Phil. 62:1--12.
Bartlett, H. 1926. "Sumatran Plants in Asahan and Karoland, With Notes on
Their Vernacular Names." Mich. Acad. ScL Arts Let., 6:1--66.
1935. "The Batak Lands of North Sumatra." Natural Appli. Sci. Bull.,
4:211--323. (University of the Philippines.)
- 1940. "History of the Generic Concept in Botany." Bull. Torrey Bot. Club,
47:319--362.
Bauhin, C. 1623. Pinax theatri botanici. Basel: Regis.
Berlin, B. 1972. "Speculations on the Growth of Ethnobotanical Nomenclature."
Lang. Soc., :63--98.
1973. "The Relation of Folk Systematics to Biological Classifications and
Nomenclature." Ann. Rev. Syst. Ecol., 4:259--271.
- 1978. "Ethnobiological Classification." In Cognition and Categorization,
ed. E. Rosch and B. Lloyd, pp. 9--26. Hillsdale, N.J.: Erlbaum.
1982a. "Predicting Discontinuities in Ethnobiological Classification." Paper
presented at the Department of Anthropology Faculty Seminar, November
22, 1982, University of California, Berkeley.
1982b. "Natural and Not-so-natural Higher-order Categories in Ethnobotanical Classification." Paper read before the Anthropology Section of the
New York Academy of Sciences, April 26, 1982, New York.
Berlin, B.; Boster, J.; and O'Neill, J. 1981. "The Perceptual Basis of Ethnobiological Classification: Evidence from Aguaruna Folk Ornithology." J.
Ethnobiol., 1:95--108.
Berlin, B.; Breedlove, D.; and Raven, P. 1973. "General Principles of Classification and Nomenclature in Folk Biology." Amer. Anthrop., 75:214--242.
- 1974. Principles of Tzeltal Plant Classification. New York: Academic.
Bianchi, H. 1957. "Tournefort et la meSdicine." In Tournefort, ed. R. Heims, pp.
97--103. Paris: Mus6um National d'Histoire Naturelle.
Blackwelder, R. 1967. Taxonomy. New York: Wiley.
Bock, H. 1539. Kreiitter Buch. S trasbourg: W. Rihelius.
1552. De stirpium. Strasbourg: W. Rihelius.
Bock, W. 1973. "Philosophical Foundations of Classical Evolutionary Taxonomy." Syst. Zool., 22:375--392.
Boyle, R. 1667. The Origine of Formes and Qualities. 2nd ed. Oxford: Davis.
Bremekamp, C. 1953. "A Re-examination of Cesalpino's Classification." Acta
Bot. Neerl., :580--593.
Bretzl, H. 1903. Botanische Forschungen des Alexanderzuges. Leipzig: Teubner.
Brown, C. 1977. "Folk Botanical Life-forms: Their Universality and Growth."
-
Amer. Anthrop., 79:317--342.

1979. "Folk Zoological Life-forms: Their Universality and Growth." Amer.
Anthrop., 81:791--817.
- 1984. Language and Living Things: Uniformities in Folk Classification and
Naming. New Brunswick, N. J. Rutgers University,
Brunfels, O. 1530-- 1536. Herbarum vivae eicones. Strasbourg: Schottu.
Buffon, G.-L. 1749. Histoire naturelle g~n~rale et particulidre, vol. I. Paris:
-
Imprimerie Royale.
Bulmer, R. 1970. "Which Came First, the Chicken or the Egg-head?" In
I~changes et communications: Mdlanges offerts d Claude Ldvi-Strauss, ed. J.
Pouillon and P. Maranda, pp. 1069--91. The Hague: Mouton,
274
SCOTT A T R A N
1974. "Folk Biology in the New Guinea Highlands." Soc. Sci. Info.,
13:9--28.
1979. "Kalam Classification of Birds." In Classifications in The& Social
Contexts, ed. R. Ellen and D. Reason, pp. 57--79. New York: Academic.
Bulmer, R., and Tyler, M 1968. "Karam Classification of Frogs." J. Polynesian
Soc., 77:333--385.
Burma, B 1949. "The Species Concept: A Semantic Review." Evolution,
3:369--373.
Cain, A. 1956. "The Genus in Evolutionary Taxonomy." Syst. Zool., 5:97--109.
- 1959. "The Post-Linnaean Development of Taxonomy." Proc. Linn. Soc.
-
London, 170:234--244.
Candolle, A.-P. 1833. "Note sur la division du r~gne v6g6tal en quatre
embranchements." Bibl. Universelle Gen~ve Sci. Arts, 18:259--268.
Carey, S. 1978. "The Child's Concept of 'Animal.'" Paper presented to the
Psychonomic Society, San Antonio, Texas.
Cassirer, E. 1906. Das Erkenntnisproblem in der Philosophie und Wissenschafi
de neueren Zeit. Berlin: Bruno Cassirer.
Cesalpino, A. 1571. Questionum peripateticorum libri quinque. Venice: Juntas.
1583. De plantis libri X VI. Florence: Marescot.
Cole, C. 1984. "Taxonomy: What's in a Name?" Nat. Hist., 93:30--34.
Cordus, V. 1561. Valerii Cordi Simesusi] Annotationes in Pedaci] Dioscoridis
Anazerabei de medica materia libros V [Historia plantarum]. Edited by C.
Gesner. Strasbourg: J. Rihelius.
Croizat, L. 1945. "History and Nomenclature of the Higher Units of Classification." Bull. Torrey Bot. Club, 72:52--75.
Cronquist, A. 1968. The Evolution and Classification of Flowering Plants. New
York: Houghton Mifflin.
Cullen, J. 1968. "Botanical Problems of Numerical Taxonomy." In Modern
Methods in Plant Taxonomy, ed. V. Heywood, pp. 175--183. New York:
Academic.
Cuvier, G. 1812. "Sur un nouveau rapprochement ii 6tablir entre les classes qui
composent le r~gne animal." Mus6um d'Historie Naturelle de Paris, Annales,
9:73--84.
Darwin, C. 1883. On the Origin of Species by Means of Natural Selection. 6th ed.
New York: Appleton.
Daudin, H. 1926--1927. Cuvier et Lamarck: Les classes zoologiques et l'id~e de
s~rie animale (1790--1830). 2 vols. Paris: Alcan.
Davis, P., and Heywood, V. 1963. Principles of Angiosperm Taxonomy. New
York: Van Nostrand.
de Beer, G., ed. 1960, 1967. "Darwin's Notebooks on Transmutation of Species."
Bull. Brit. Mus. (Nat. Hist.), 2:27--200; 3:129--176. (Notebooks B, C, D, E.)
Descartes, R. 1907 [1637]. Discours de la m~thode. Paris: Cerf.
Diamond, J. 1966. "Zoological Classification of a Primitive People." Science,
15:1102--4.
- 1972. The Avifauna of the Eastern Highland of New Guinea. Cambridge,
Mass.: Nuttall Ornithological Club.
Dioscorides. 1959. The Greek Herbal of Dioscorides, Illustrated by a~Byzantine
A.D. 712, Englished by John Goodyer A.D. 1655. Ed. R. Gunther. New York:
Hafner.
Diver, C. 1940. "The Problem of Closely Related Species Living in the Same
Area." In The New Systematics. ed. T. Huxley, pp. 301--328. Oxford:
Clarendon.
-
275
Dobzhansky, T. 1937. Genetics and the Origin of Species. New York: Columbia
University.
Dodoens, R. 1583. Stirpium historiae. Antwerp: Plantini.
Dorolle, M. 1929. Introduction to C~salpin, questions p~ripat~ticiennes. Paris:
Alcan.
Dougherty, J. 1979. "Learning Names for Plants and Plants for Names." Anthrop.
Linguistics, 21:298--315.
Dughi, R. 1957. "Tournefort dans l'histoire de la botanique." In Tournefort, ed.
R. Heim, pp. 131--185. Paris: Mus6um National d'Histoire Naturelle.
Durkheim, E., and Mauss, M. 1963 [1903]. Primitive Classification. Trans. R.
Needham. Chicago: Chicago University.
Dwyer, P. 1976a. "An Analysis of Rofaifo Mammal Taxonomy." Amer. Ethnol.,
3:425--445.
1976b. "Beetles, Butterflies, and Bats: Species Transformation in New
Guinea Folk Classification." Oceania, 14:188--205.
Emmart, E. 1940. The Badianus Manuscript: An Aztec Herbal. Baltimore:
(Originally composed in 1552.)
Farber, P. 1982a. "The Transformation of Natural History in the Nineteenth
Century." J. Hist. BioL, 15:145--152.
- 1982b. The Emergence of Ornithology as a Scientific Discipline, 1760-1850. Dordrecht: Reidel.
Feyerabend, P. 1970. "Against Method: Outline of an Anarchistic Theory of
Knowledge." Minnesota Stud. Phil, Sci., 4:17-- 130.
Foucault, M. 1970 [1966[. The Order of Things. London: Tavistock.
Friedberg, C. 1970. "Analyse de quelques groupements de v6g6taux comme
introduction h l'~tude de la classification botanique bunaq." In ~changes et
communications: Mdlanges offerts d Claude Ldvi-Strauss, vol. II, ed J. Pouillon
and P. Maranda, pp. 1092--1130. The Hague: Mouton.
1972. "l~16ments de botanique bunaq." In Langues et techniques." Nature et
societY, vol. II, Approche ethnologique et naturaliste, ed. J. Barrau, pp. 375-393. Paris: Klincksieck.
- 1982. Les Bunaq de Timor et les Plantes, vol. IV MUK GUBUL NOR "La
Chevelure de la Terre." Th~se d'l~tat, Universit6 de Paris V.
Fuchs, L. 1542. De historia stirpium. Basel: Insengrin.
- 1543. Neu Kreiiterbuch. Basel: Insengrin.
Galileo, G. 1623. II Saggiatore. Rome: Mascardi.
Gesner, C. 1584. Epistolarum medicinalium. Wittenberg: Gronenberg.
- 1972. Conradi Gesneri historia plantarum. Zurich: U. Graf.
Gilbert, N. 1963. "Galileo and the School of Padua." J. Hist. Phil., 1:223--231.
Giimour, J., and Waiters, S. 1964. "Philosophy and Classification." In Vistas in
Botany, vol. IV, Recent Researches in Plant Taxonomy, ed. W. Turrill, pp.
1--22. Oxford: Pergamon.
Gotthelf, A. 1976. "Aristotle's Conception of Final Causality." Rev. Metaphys.,
30:226--254.
Greene, E. 1983. Landmarks of Botanical History, 2 vols. Stanford: Stanford
University.
Guy6not, E. 1941. Les sciences de la vie au XVII ~ et XVIII e sidcles. Paris: Albin
Michel.
Harvey, W. 1981 [1651]. Disputations Touching the Generation of Animals [De
generatione animalium]. Trans. G. Whitteridge. Oxford: Blackwell.
Hays, T. 1976. "An Empirical Method for the Identification of Covert Categories
in Etlmobiology." Amer. EthnoL, 3:489--507.
-
276
scott
ATRAN
1983. "Ndumba Folk Biology and General Principles of Ethnobotanical

Classification and Nomenclature." Amer. Anthrop., 85:592--611.
Hodge, M. 1985. "Darwin, Species, and the Theory of Natural Selection." Paper
presented to Colloque International: "Histoire du Concept d'Esp6ce dans
les Sciences de la Vie," Fondation Singer-Polignac, Paris, May 28--31,
1985.
Hull, D. 1965. "The Effect of Essentialism on Taxonomy -- Two Thousand
Years of Stasis. Part 1." Brit. J. Phil Sci., 15:314--326.
Hunn, E. 1975a. "Cognitive Processes in Folk Ornithology: The Identification of
Gulls." Language-Behavior Research Laboratory, Working Paper No, 42,
University of California, Berkeley.
1975b. "A Measure of the Degree of Correspondence of Folk to Scientific
Biological Classification." Amer. Ethnol., 2:309--327.
1976. "Toward a Perceptual Model of Folk Biological Classification."
Amer. Ethnol,, 3:508--524.
Inger, R. 1958. "Comment on the Definition of Genera." Evolution, 2:370-384.
lnglis, W. 1966. "The Observational Basis of Homology." Syst. Zool., 15:219-228.
Jacob, F. 1973 [1970]. The Logic of Life. Trans. B. Spillman. New York:
Pantheon.
,lacobs, M. 1980. "Revolutions in Plant Description." In Liber gratulatorius in
honorem H. C. D. De Wit, ed. J. Arends, G. Boelema, C. de Groot, and A.
Leeuwenberg, pp. 155--181. Wageningen: H. Veenman and Zonen.
Jardine, N. 1969. "A Logical Basis for Biological Classification." Syst. Zool.,
8:37--52.
1976. "Galileo's Road to Truth and the Demonstrative Regress." Stud.
Hist. Phil, Sci. 7:277--318.
Jaume Saint-Hilaire, ,l. H. 1805. Exposition des families naturelles. 2 vols. Paris:
Treuttel and Wiirtz.
Joly, R. 1968. "La biologie d'Aristote." Rev. philos., 158:219--253.
.lung, J. 1747 [1662]. Opuscula botanico-physica. Ed. M. Fogel and J. Vagetius.
Coburg: Ottonis. (The Isagoge phytoscopica was originally complied in 1678-1679 and Deplantis doxoscopiae in 1662.)
- 1982. Praelectiones physicae. Historisch-Kritische Edition, ed. Christoph
Meinel. G6ttingen: Vandenhoeck and Ruprecht.
Jussieu, A.-L. 1789. Genera plantarum. Paris: Herissant.
K~istner, H. 1896. "Pseudo-Dioscorides De herbis femininis." Hermes, 3:578-636.
Keil, F. 1979. Semantic and Conceptual Development." An Ontological Perspective. Cambridge, Mass.: Harvard University.
Kesby, J. 1979. "The Rangi Classification of Animals and Plants." In Classifications in Their Social Contexts, ed. R. Ellen and D. Reason, pp. 33--56. New
York: Academic.
Lamarck, J.-B. 1758. M~moire sur les classes les plus convenables ti ~tablir parmi
les v~g~taux. M~moires de l'Acad~mie des Sciences, ann6e 1758, pp. 437-453.
1809. Philosophie zoologique. Paris: Dentu.
Landgraf, E. 1928. "Ein friihmittelaltericher Botanicus." Kyklos, 1:3--36.
Lenox, J. 1980. "Aristotle on Genera, Species, and 'The More and the Less.'" J.
Hist. Biol., 13:321--346.
Li, H.-L. 1974. "Plant Taxonomy and the Origin of Cultivated Plants." Taxon,
23:715--724.
-
277
Linnaeus, C. 1737a. Genera plantarum. Leiden: Wishoff.

- 1737b. Critica botanica. Leiden: Wishoff.
1751. Philosophia botanica. Stockholm: Kiesewetter.
1787. Philosophia botanica. 4th ed. Geneva: Pietre and Dellamoli~re.
- 1790 [1760]. Disquisitio de sexu plantarum. In Amoenitates academicae,
vol. X, pp. 100--131. Erlangen: Schreber. (Originally presented as the prize
essay of the Imperial Academy of St. Petersburg in 1760.)
- 1972 [1744[. Discours sur l'accroissement de la terre habitable [Oratio de
Telluris habitabilis incremento[. In L'dquilibre de la nature, trans. B. Jasmin,
pp. 29--55. Paris: Vrin. (Originally published in Latin in 1744, subsequently
in the Amoenitates academicae, vol. II.)
Lloyd, G. 1968. Aristotle: The Growth and Structure of HIS Thought. Cambridge:
Cambridge University.
Locke, J. 1848 [1689]. An Essay Concerning Human Understanding. London:
Tegg.
MacArthur, R. 1972. Geographical Ecology." Patterns in the Distribution of
Species. New York: Harper & Row.
Macnamara, J. 1982. Names for Things:A Study of Human Learning. Cambridge,
Mass.: MIT.
Magnol, P. 1689. Prodromus historiae generalisplantarum. Montpellier: Pech.
Mayr, E. 1969. Principles of Systematic Zoology. New York: McGraw-Hill.
- 1982. The Growth of Biological Thought. Cambridge: Harvard University.
Meinel, C. 1984a. In physicis futurum saeculum respicio: Joachim Jungius und
die NaturwissenschaJ~liche Revolution des 17 Jahrhunderts. G6ttingen: Vandenhoeck and Rupreeht.
- 1984b. Der handschriftliche Nachlass von Joachim Jungius in der Staatsund Universitiitsbibliothek Hamburg. Katalog. Stuttgart: Ernst Hauswedell.
1985. "R6volution scientifique et exp6rience en chimie: m6thodologie et
r6forme chez Joachim Jungius vers 1630." Paper presented to Coiloque
International: "H61bne Metzger," Centre International de Synth~se, Paris, May
21--23, 1985.
Meyer, E. 1854--1857. Geschichte derBotanik. 4 vols. K6ningsberg: Borntr~iger.
Milt, B. 1936. "Conrad Gessner's 'Historia plantarum' (Fragmenta relicta)."
Vierteljahrs. naturforsch. Ges. Ziirich, 81:285--291.
Morison, R. 1669. Praeludia botanica. London.
- 1672. Plantarum umbelliferarum distributio nova. Oxford: Sheldonian.
1680--1699. Plantarum historiae universalis Oxiensis. Vols. II--III.
Oxford: Sheldonian.
Morton, A. 1981. History of BotanicalScience. New York: Academic.
Needham, J. 1986. Science and Civilisation in China. Vol. VI(1), Botany.
Cambridge: Cambridge University.
Owen, R. 1866. The Anatomy of Vertebrates. Vol. I, Fishes and Reptiles. London:
Longman, Green.
Paso y Trancoso. F. 1886. "La botanica entre los Nahuas." Anales Mus. Nac.
Mexico, 3.
Pline [Pliny]. 1829. Histoire naturelle de Pline [Historiarum Mundi]. 20 vols.
Paris: Panckoucke.
Quine, W. 1966. "Three Grades of Modal Involvement." In W. Quine, The Ways
of Paradox, pp. 174--184. New York: Random House.
Randall, J. 1940. "The Development of Scientific Method in the School of
Padua." J. Hist. Ideas, 1:177--206.
Raven, C. 1942. John Ray Naturalist, His Life and Works. Cambridge:
Cambridge University.
-
278
scott
ATRAN
Raven, P. 1976. "Systematics and Plant Population Biology." Syst. Bot., :284-316.
Raven, P.; Berlin, B.; and Breedlove, D. 1971. "The Origins of Taxonomy."
Science 174:1210--13.
Ray, J. 1660. Catalogous plantarum circa Cantabrigiam nascentium. Cambridge:
Field.
1682. Methodus plantarum nova. London: Faithorne and Kersey.
1686. Historia plantarum. London: Smith and Walford.
- 1691. The Wisdom of God Manifested in the Works of Creation. London:
Smith.
- 1693. Synopsis methodica animalium quadrupedum & serpentini. London:
Smith and Walford.
- 1696a. Synopsis methodica stirpium Britanicarum. London: Smith and
Walford.
- 1696b. De variis plantarum methodus dissertatio brevis. London: Smith
and Walford.
1703. Methodus plantarum emendata et aucta. London: Smith and
Walford.
- 1848. The Correspondence of John Ray. Edited by E. Lankester. London:
Ray Society.
Regan, C. 1926. "Organic evolution." Presidential Address, Section D. British
Association for the Advancement of Science. 1925.
Riddle, J. 1981. "Pseudo-Dioscorides' Ex herbis femininis and Early Medieval
Medical Botany." J. Hist. Biol., 4:43--81.
Rivinus, A. [Bachmann]. 1690. Introductio generalis in rein herbariam. Leipzig:
Giinther.
Sachs, J. 1890 [1875]. History of Botany (1530--1860). Trans. H. Garnsey.
Oxford: Clarendon.
Scheffer, W. 1958. Seals, Sea Lions, and Walruses. Stanford: Stanford Univeristy.
Simpson, G. 1961. Principles of Animal Taxonomy. New York: Columbia
University.
Singer, C. 1927. "The Herbal in Antiquity." J. Hellenic Stud., 47:1--52.
- 1959. A History of Biology. New York: Abelard-Schuman.
Skulsky, H. 1968. "Paduan Epistemology and the Doctrine of the One Mind." J.
Hist. Phil., 6:341--361.
SIoan, P. 1972. "John Locke, John Ray, and the Problem of the Natural System."
J. Hist. Biol., 5:1--53.
Sperber, D. 1975. "Pourquoi les animaux parfaits, les hybrides et les monstres
sont-ils boris a penser symboliquement?" L'Homme, 5:5--34.
Sprague, T. 1928. "The Herbal of Otto Brunfels." J. Linn. Soc. London, 48:79-124.
1940. "Taxonomic Botany, with Special Reference to the Angiosperms." In
The New Systematics, ed. T. Huxley, pp. 435--453. Oxford: Clarendon.
Sprague, T., and Sprague M. 1939. "The Herbal of Valerius Cordus." J. Linn.
-
Soc. London, 52:1-- 113.

Sprengel, K. 1817-- 1818. Geschichte der Botanik. 2 vols. Altenberg: Brockhaus.
Stearn, W. 1957. "An Introduction to the Species plantarum and Cognate
Botanical Works of Carl Linnaeus." In Species plantarum. A Facsimile of the
first edition, 1753. vol. I, pp. 1--176. London: Ray Society.
Stevens, P. 1984a. "Metaphors and Typology in the Development of Botanical
Systematics 1690--1960, or The Art of Putting New Wine in Old Bottles."
Taxon, 33:169--211.
279
1984b. "Haiiy and A.-P. de Candolle: Crystallogrpahy, Botanical Systematics, and Comparative Morphology, 1780--1840." J. Hist. BioL 7:49-82.
Strathern, M. 1696. "Why is the Pueraria a Sweet Potato?" Ethnology, 8:189-198.
Stross, B. 1973. "Acquisition of Botanical Terminology by Tzeltal Children." In
Meaning in Mayan Languages, ed. M. Edmonson, pp. 107--141. The Hague:
Mouton.
Taylor, P. 1978--1979. "Preliminary Report on the Ethnobiology of the
Tobelorese of Hamalhera, North Moluccas." Majalah llmu-ilmu Sastra
Indonesia, 8:215--229.
1984. "Covert Categories' Reconsidered: Identifying Unlabeled Classes in
Tobelo Folk Biological Classification." J. Ethnobiol., 4:105-- 122.
Theophrastus. 1968. Enquiry into Plants [Historia plantarum]. Trans. A Hort.
London: Heinemann.
Tournefort, J.-P. 1694. l~l~mens de botanique. Paris: Imprimerie Royale.
-1697. De optima methodo instituenda in re herbaria. Epistola. In quiT
respondetur Dissertationi D. Raii de variis plantarurn methodus. Paris: Jardin
du Roi.
- 1708. Materia medica, or a Description o f Simple Medicines Generally
Used in Physick . . . Their Operating and Acting upon Human Bodies According to the Principles o f the New Physiology, Chymistry and Mechanism.
London: Bell. (Lessons given at the Jardin du Roi, Paris, and first published in
English.)
1717. Relation d'un voyage au Levant. 3 vols. Paris: Imprimerie Royale.
1719. Institutiones rei herbariae. 3rd ed. Paris: Imprimerie Royale.
Vines, S. 1913. "Robert Morison and John Ray." In Makers of British Botany, ed.
F. Oliver, pp. 8--43. Cambridge: Cambridge University.
Voigts, L. 1978. 'q?he Signficance of the Name Apuleius to the Herbarium
Apulei." Bull. Hist. Med., 52:214--227.
Walters, S. 1961. "The Shaping of Angiosperm Taxonomy." New Phytologist,
60:70--84.
Wellisch, H. 1975. "Conrad Gessner: A bio-bibliography." J. Soc. Bibliog. nat.
Hist., 7:151--247.
Wharburton, F. 1967. "The Purposes of Classification." Syst. Zool., 6:241--245.
Wilkins, J. 1668. Essay Towards a Real Character and Artificial Language.
London: The Royal Society.
Zabarella, G. 1608. lacpbi Zabarellae Patavini opera logica. Frankfurt: Zetzneri.

Origin of The Species and Genus Concepts

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Origin of The Species and Genus Concepts

Uploaded by

Copyright:

Available Formats

Origin of the Species and Genus Concepts:

In his dedicatory epistle to the founding work of systematic

do better - - the unspoiled savage of Montaigne notwithstanding.

Origin of the Species and Genus Concepts

appear that herbals tend to make their appearance at a rather

unknown) natures with propensities responsible for the readily

Origin of the Species and Genus Concepts

ranean plant and animal types depicted in ancient sources. The

Fuchs, o n e o f the G e r m a n "fathers of b o t a n y , " r e m a r k s that

4. Tournefort well understood that ancient knowledge was more thorough

Origin of the Species and Genus Concepts

tunate but pervasive ignorance of the anthropological foundations

correspond, within predictable limits, to those species of the field

Origin of the Species and Genus Concepts

T h e p r i m a r y o p e r a t i o n in the classification o f living kinds,

Origin of the Species and Genus Concepts

organisms into 'species"' (Jardine 1969:37). This basic partitioning

complication of perceptually aberrant mutants, incomplete fossil

Similarly, in zoology "the basic or one might say primitive step

Origin of the Species and Genus Concepts

appears to be true of folkzoological taxonomies (Berlin, Breedlove

is, in the context of a search of worldwide order -- an order

Origin of the Species and Genus Concepts

ficient, are not always necessary. T h e necessary condition is an

ecological distinction that marks a m o r p h o - g e o g r a p h i c a l gap in the

8. In cases of species recognition for asexual organisms, Mayr acknowledges

Origin of the Species and Genus Concepts

foreign species is truly exotic, such a strategy inevitably fails. Thus,

Origin of the Species and Genus Concepts

"true" (bac'il); hence, it too is nominally binomial in structure

specieme has apparently achieved conceptual equality with the

Origin of the Species and G e n u s C o n c e p t s

ship of genus to species can be established between Bappelen and

Origin of the Species and Genus Concepts

Brunfels had earlier noted the existence of Kiicheschell but chose

relationships with other plants; and the mode by which ferns

Origin of the Species and Genus Concepts

22. Gesner did, in fact, compile a collect,un of some 1500 reasonably

initially a p p e a r to the historian of biology unfamiliar with folkbiology.

Origin of the Species and Genus Concepts

per section, whereas Bauhin sometimes names as many as twenty

polynomial structures that begin with the binomial indicating the

Origin of the Species and Genus Concepts

pirical problems quite different from those envisaged by Aristotle,

Origin of the Species and G e n u s C o n c e p t s

not in themselves, but vicariously t h r o u g h their offspring, which,

30. Eidos and genos appear as interchangeable terms in Aristotle's oeuvre,

Origin of the Species and Genus Concepts

For Aristotle, as for Cesalpino, the form (eidos), or soul

k n o w l e d g e . O n A r i s t o t l e ' s a c c o u n t , though, to the e x t e n t that a

Origin of the Species and Genus Concepts

primarily to their morphological manifestations. Moreover, because Aristotelian

engenders like, according to nature and of the same species.

Origin of the Species and Genus Concepts

terizes ancient Hebrews, Chinese, and Mesoamericans, as it does

36. Gesner may have preceded Cesalpino in recommending the experimental

Origin of the Species a n d G e n u s C o n c e p t s

c o l o r are particularly subject to variation a n d c a n b e used only as

A n d if there were need of a third "vegetative substance," it

cation is sufficiently detailed to make reference to other parts

38. Tournefort (1694:17) acknowledges that Gesner originally suggested the

Origin of the Species and Genus Concepts

lack of tendrils. In the end, though, all secondary divisions down to